Clinical features and long-term prognosis of trochlear headaches

Transcription

1 European Journal of Neurology 2013 doi: /ene Clinical features and long-term prognosis of trochlear s J. H. Smith a, J. A. Garrity b and C. J. Boes c a Department of Neurology, University of Kentucky, Lexington, KY; b Department of Ophthalmology, Mayo Clinic, Rochester, MN; and c Department of Neurology, Mayo Clinic, Rochester, MN, USA Keywords: chronic daily, migraine, neuroophthalmology, ocular movements, secondary disorders, trochlea, trochleitis Received 13 September 2013 Accepted 21 October 2013 Background and purpose: Trochlear s are a recently recognized cause of, of which both primary and inflammatory subtypes are recognized. The clinical features, long-term prognosis and optimal treatment strategy have not been well defined. Methods: A cohort of 25 patients with trochlear seen at the Mayo Clinic between 10 July 2007 and 28 June 2012 were identified. Results: The diagnosis of trochlear was not recognized by the referring neurologist or ophthalmologist in any case. Patients most often presented with a new daily from onset (n = 22, 88%). The most characteristic syndrome was reported as continuous, achy, periorbital pain associated with photophobia and aggravation by eye movement, especially reading. Individuals with a prior history of migraine were likely to have associated nausea and experience trochlear migraine. Amongst individuals with trochleitis, 5/12 (41.6%) had an identified secondary mechanism. Treatment responses were generally, but not invariably, favorable to dexamethasone/lidocaine near the trochlea. At a median follow-up of 34 months (range 0 68), 10/25 (40%) of the cohort had experienced complete remission. Conclusions: Trochlear s are poorly recognized, have characteristic clinical features, and often require serial to optimize the treatment outcome. The identification of trochleitis should prompt neuroimaging to look for a secondary cause. Introduction The trochlea is a saddle-like cartilaginous structure located in the superomedial orbit, which contains the tendon of the superior oblique muscle. Histologically, the trochlea is surrounded by a synovial membrane, analogous to the structure of a joint [1]. The trochlea is innervated by an ophthalmic nerve branch and is capable of generating pain in the setting of trochleitis, which is most often idiopathic, but may develop secondary to autoimmune connective tissue disorders [2]. Trochleitis is characterized by swelling and tenderness of the trochlea, and aggravation by vertical ductions. Swelling may be observed clinically or documented on orbital imaging studies [2]. A non-inflammatory condition, termed primary trochlear (PTRH), has also been described [3]. Treatment of all trochlear s is presumed to require local steroid injection in the vicinity of the trochlea, and is Correspondence: J. H. Smith, Department of Neurology, University of Kentucky, 740 S. Limestone, L445, Lexington, KY 40503, USA (tel.: ; fax: ; jonathan.smith@uky. edu). generally thought to be associated with a positive response. Unfortunately, only a very limited number of cases have been reported in the literature [3 5]. Therefore, information regarding the clinical features, treatment and prognosis is limited. To help clarify these issues, our experience with trochlear in 25 previously unreported cases is reported. Methods Prior to the initiation of the study, Mayo Clinic Institutional Review Board approval was obtained. The procedural records of one of the investigators (JAG) was searched from 10 July 2007 to 28 June 2012 to identify all adult (age 18) patients who had received a trochlear injection. The electronic medical records were manually abstracted to identify individuals where a diagnosis of a trochlear could be definitively assigned and adequate clinical documentation was available. In our paper, the general term trochlear is used to describe any referable to the troch- 1

2 2 J. H. Smith, J. A. Garrity and C. J. Boes lear apparatus. Given the lack of validated criteria, specific diagnoses of PTRH and trochlear migraine were assigned based on previously suggested definitions [3,5]. Trochlear migraine refers to a trochlear which then triggers a secondary migraine attack [2,5]. The diagnostic criteria for secondary to trochleitis were used as suggested by the newly published International Classification of Headache Disorders, 3rd edition (online beta version) [6]. A diagnosis of secondary trochleitis was assigned if there was clinical, radiographic and/or histopathological evidence for orbital localization of an inflammatory or neoplastic process. The diagnosis of either trochleitis or PTRH could be supported by aggravation and/or reproduction of pain by action of the superior oblique muscle. Patients were excluded if an alternative diagnosis for the was established, such as a periorbital cranial neuralgia, a trigeminal autonomic cephalalgia or carotid-cavernous fistula. Acuity of onset was summarized as acute if the was daily from onset, subacute if daily within 1 month of onset, and insidious if daily within 1 year of onset. Due to the retrospective nature of the study, treatment efficacy was summarized as complete, partial or ineffective based on the information available in the medical record. If the patient reported a quantitative indicator of pain relief, this information was specifically recorded. Results Over a 5-year period of chart review, 25 individuals were identified with sufficient clinical characterization for study inclusion (Table 1). Patients were generally female (n = 20, 80%), with a median age at diagnosis of 46 (range 18 77). Patients had a median time from symptom onset to diagnosis of 6.7 months (range 2 weeks to 10 years). The diagnosis of a trochlear had not been made in any patient prior to referral, despite all patients having been seen previously by either a neurologist or ophthalmologist prior to presentation. The most common mis-diagnoses prior to presentation were chronic migraine (n = 13), new daily persistent (n = 5), no diagnosis (n = 4), hemicrania continua (n = 2) and atypical facial pain (n = 1). All patients with a prior diagnosis selfreported a new superimposed pain syndrome, which was very apparent to all patients as distinct. Antecedent events were only occasionally noted by patients. The most concrete association was in case 3, where a new daily trochlear began immediately following orbital surgery for removal of an optic nerve sheath meningioma. One individual developed the 6 months following removal of an orbital dermoid tumor (case 5), 1 month following orbital decompression for Graves ophthalmopathy (case 6), 3 months post-partum (case 11) and 1 month following a Roux-en-Y gastric bypass (case 13). No enlargement of the superior oblique muscle had been radiographically noted in case 6. There were no complications reported in review of any of the above mentioned procedures. Trochlear s had an acute, daily from onset, presentation in the majority of cases (n = 22, 88%), being subacute, daily within 1 month, in the remainder. A final diagnosis of PTRH (n = 13, 52%) was made slightly more often than trochleitis (n = 12, 48%). Bilateral involvement was seen in 8/13 (61.5%) cases of PTRH and in 4/12 (33.3%) cases of trochleitis. In all bilateral cases this always occurred sequentially, and within 1 year of onset. Amongst patients with trochleitis, five (41.6%) had an identified secondary mechanism. These included incomplete Behcet s (n = 1), idiopathic Tolosa Hunt (n = 1), granulomatosis polyangiitis (GPA) (n = 2) and orbital lymphoma (n = 1). All five cases of secondary trochleitis underwent biopsy of an orbital mass located near the trochlea. In case 11, the trochlear had been the presenting symptom of GPA. Secondary trochleitis was most often bilateral, but was observed unilaterally in one individual (case 11). The location of the trochlear was most often focused at the medial, orbit or forehead (n = 9, 36%), with radiation into the forehead, temple, periorbit or retro-orbitally. The pain was reported as continuous (n = 25, 100%), and achy, dull or pressure-like (n = 19, 76%). The average intensity as rated on a numerical rating scale was severe (7 10) in seven (35%), moderate (4 6) in 10 (50%) and mild (0 3) in three (15%) cases. The most common associated symptoms were photophobia (n = 15, 68.1%) and binocular diplopia (n = 10, 45.4%). Four out of the five patients who reported associated nausea also had comorbid migraine. The was characteristically aggravated by eye movements (especially reading) by 18 (75%) patients. Amongst individuals with a prior diagnosis of migraine (n = 7), six were considered to have attacks of trochlear migraine following the onset of the new syndrome. These s were reported to be ipsilateral to the trochlear pain. Diagnostic evaluation consisted of either a magnetic resonance imaging (MRI) of the head or a computed tomography (CT) study of the orbits in all patients except one (case 13). Three patients were found to have an imaging abnormality directly involving the

3 Prognosis of trochlear s 3 Table 1 Clinical features of patients with trochlear no. Age/gender Diagnosis/laterality Acuity of onset Continuous or episodic Location Radiation Quality/ intensity Associated symptoms Aggravating factors Relieving factors Relevant comorbidity 1 55/F PTRH, TM/R Acute Continuous Orbit None Pressure/6 NA, PHT Reading None CMA 2 29/M Trochleitis, TM/Bil Acute Continuous Orbit, medial None Throb/6 NA, PHT, DIP EM None EMO 3 77/F PTRH/R Acute Continuous Medial None Ache NR None None Left optic nerve meningioma 4 34/F PTRH, TM/Bil Acute Continuous Periorbital Holocephalic Throb NA, PHT, DIP None None EMO, IED 5 32/M PTRH/Bil Acute Continuous Retroorbital Temple Pressure/6 NR EM NR Right orbital dermoid 6 57/F PTRH/Bil Acute Continuous Eye NR NR/7 DIP EM NR Graves ophthalmopathy s/p decompression 1 month prior to onset 7 38/F Trochleitis, Acute Continuous Orbit None Ache/7 NA, DIP EM None CMO incomplete Behcet s/bil 8 36/F Trochleitis/ Tolosa Hunt/Bil Acute Continuous Periorbital Forehead Pressure/7 NA, DIP EM Eye patch 9 63/M Trochleitis/R Acute Continuous Periorbital None Ache/8 NR None None History of encephalitis in childhood, unknown cause 10 18/F Trochleitis, TM/L Acute Continuous Medial None NR NR NR NR EMO 11 56/F Trochleitis, GPA/L Acute Continuous Eyebrow None Ache/2 PHT, DIP EM None GAD 12 42/F PTRH/R Acute Continuous Medial Forehead Pressure/5 NR EM None 13 71/F PTRH/Bil Acute Continuous Periorbital None Dull/3 PHT EM Warmth History of gastric bypass 14 65/F Trochleitis/Bil Acute Continuous Retroorbital Eyebrow Pressure, sharp/4 PHT, DIP EM None 15 57/F PTRH/R Acute Continuous Retroorbital None Pressure/4 PHT EM Warmth 16 40/F Trochleitis/R Acute Continuous Medial forehead 17 74/F PTRH/Bil Acute Continuous Medial 18 58/F PTRH/Bil Acute Continuous Medial forehead 19 46/F PTRH, TM/Bil Acute Continuous Medial Retro-orbital Pressure/4 PHT, DIP EM None Forehead, temple Pressure/6 PHT EM Warmth EMO, depression None Ache PHT, PHN Touch None Forehead, bridge of nose Pressure/4 PHT EM None Depression 20 23/F Trochleitis, GPA/L Acute Continuous Eyebrow Orbit Throb/7 PHT, DIP EM None GPA 21 48/M Trochleitis/L Acute Continuous Orbit None Pressure/2 None EM None ETTH

4 4 J. H. Smith, J. A. Garrity and C. J. Boes Table 1 (Continued ) Relevant comorbidity Relieving factors Aggravating factors Associated symptoms Quality/ intensity Continuous or episodic Location Radiation Acuity of onset no. Age/gender Diagnosis/laterality 22 35/F Trochleitis, TM/L Subacute Continuous Periorbital Eyebrow Pressure PHT, DIP EM None EMO GBM, s/p WBRT (1 year before onset) +Radiationinduced retinopathy and optic neuropathies Forehead Pressure/6 PHT None Eye closure 23 48/M PTRH/L Acute Continuous Medial None None History of anticardiolipin antibody Subacute Continuous Periorbital Hemicranial/neck Throbbing/7 PHT, blurry vision 24 41/F Trochleitis, orbital MALT lymphoma/l 25 39/F PTRH/Bil Subacute Continuous Retroorbital None Dull/7 None EM None None Bil, bilateral; CMA, chronic migraine with aura; CMO, chronic migraine without aura; DIP, diplopia; EM, eye movements; EMO, episodic migraine without aura; ETTH, episodic tension-type ; GAD, generalized anxiety disorder; GBM, glioblastoma multiforme; GPA, granulomatosis with polyangiitis; IED, intermittent explosive disorder; L, left; MALT, mucosa-associated lymphoid tissue; NA, nausea; NR, not reported; PHN, phonophobia; PHT, photophobia; R, right TM, trochlear migraine; WBRT, whole brain radiotherapy. Intensity refers to what the patient reported as average on a numerical rating scale from 0 to 10. trochlear apparatus (Fig. 1). In the patient with Tolosa Hunt (case 8), no radiographic abnormalities were visualized in the orbit. Two other patients were noted to have radiographic enlargement of the lacrimal glands (cases 11 and 16). Clinical examination was generally unremarkable except for trochleodynia (all cases), clinically apparent trochlear edema (all cases of trochleitis) and findings attributable to neuroophthalmic comorbidities (Table 1). The treatment outcomes are summarized in Table 2. Overall, at a median follow-up of 34 months (range 0 68), 10 patients had reported complete remission, 10 continued to have either persistent or recurrent trochlear but with significantly improved pain levels and four had not experienced any substantial improvements at all. Treatments with standard migraine preventive medications were generally ineffective (Table 2). Complete remission was achieved with immunotherapy (n = 2) or external beam radiation (n = 1) in cases of secondary trochleitis. Trochlear were performed in a non-standardized way with a single injection, generally containing 2 3 mg of dexamethasone, with or without triamcinolone, and lidocaine. When reported, responses occurred either immediately (n = 4) or within 7 days (n = 5). Following the injection, patients reported complete resolution of trochlear pain (n = 10), partial improvement (n = 12) or no effect (n = 8). When effective, the reported duration of effect was of the order of weeks to months (range 12 h to 12 months) (Table 2). The median number of per patient was three (range 1 12). Amongst the 17 individuals receiving repeat several patterns of response were seen. Most commonly, an initially effective injection would continue to have either a similar (n = 6) or increased (n = 4) benefit on subsequent. Of patients with an initially ineffective first injection, subsequent were either ineffective (n = 5) or subsequently effective (n = 2). Overall, the only adverse event noted was injection site bruising in one case. Discussion In our specialty practice, trochlear s were very poorly recognized by referring neurology and ophthalmology physicians. The 25 patients in our cohort presented with a fairly characteristic clinical syndrome of moderately severe, achy, periorbital (especially medial) pain, variably associated with photophobia, and aggravation by eye movements (especially reading). Individuals with a prior history of migraine were likely to have associated nausea and experience trochlear migraine as part of their clinical course. It is notable that the daily from onset

5 Prognosis of trochlear s 5 (a) (b) Figure 1 (a) Coronal computed tomography (CT) of the orbits showing calcification of the right trochlear tendon (solid arrow) and thickening of the left trochlea tendon (broken arrow) (case 21). (b) Magnetic resonance imaging (MRI) of the orbits showing gadolinium uptake at the left trochlear tendon (arrow) (case 21). (c) Axial T2-weighted fluid attenuated inversion recovery (FLAIR) MRI of the head showing diffuse left orbital involvement by GPA (arrow) (case 20). (d) Axial CT of the head with contrast showing lymphomatous involvement of the left trochlear tendon (arrow) (case 24). (c) (d) presentation could be easily confused with new-daily persistent, which is generally thought of as extremely difficult to treat [7]. However, the prevalence of undiagnosed trochlear in patients with presumed new-daily persistent has not been studied. In our cohort, patients were almost as likely to receive a diagnosis of PTRH as trochleitis. However, the distinction had important implications, as 5/12 (41.6%) patients with trochleitis (clinically apparent trochlear edema) had an identified secondary mechanism. It is therefore concluded that all individuals with trochleitis should undergo diagnostic imaging, but CT versus MRI cannot be recommended based on our limited experience. Dedicated orbital imaging is likely to be important, on the basis that the trochlea is often not visualized on routine head imaging. Based on our retrospective data, dexamethasone/lidocaine of the trochlea appeared to be a generally efficacious strategy for many patients. As many patients reported incremental success (and at times remission) with serial, at least a second round of injection is recommended for patients who are initially non-responders. This is especially important as many of these patients do not seem to respond robustly to typical migraine preventive treatments. In cases of secondary trochleitis, treatments directed at the diseasespecific process were generally efficacious. The reasons for treatment success in some patients but not others are not known. An alternative diagnostic consideration amongst refractory cases is idiopathic ophthalmodynia, which may represent a topographically restricted form of persistent idiopathic facial pain [8]. The largest case series of trochleitis was provided by Tychsen et al. in 1984, who reported on 13 patients with a syndrome of subacute onset inflammatory trochleodynia [4]. Ages ranged from 27 to 69, and 9/13 were women. Three patients developed recurrences over a mean 8 months of follow-up. Two previously anophthalmic patients underwent excisional biopsy, demonstrating perivascular lymphocytic infiltration involving not only the tendon but also extending into the superior oblique muscle itself. It was hypothesized that the condition represented a localized form of orbital pseudotumor. In our cohort, neuroimaging abnormalities were confined to the trochlear tendon in idiopathic cases and never involved the superior oblique muscle. Limitations of our study included possible selection bias, as consecutive cases were not included, and lack of a standardized approach to evaluation and treatment. Strengths of our study include the relatively large number of patients, long-term follow-up data of treatment outcomes, and fairly complete clinical data for the cohort. In conclusion, the diagnosis of a trochlear should be considered especially in patients presenting with a new daily eye pain, aggravated by eye movements, especially reading. Our data provides a new perspective on long-term treatment outcomes in this poorly recognized, but important, syndrome.

6 6 J. H. Smith, J. A. Garrity and C. J. Boes Table 2 Treatment outcomes of patients with trochlear no. Prior treatments other than trochlear injection Efficacy of treatments other than trochlear Total number of trochlear Injected material (total volume 1 ml for all cases) Onset of action of Outcome of Time from diagnosis to (months) Outcome at 1 a. Indomethacin 50 mg TID b. Botulinum toxin A, 150 units 9 1 c. Melatonin 3 mg a. Partial b. Partial c. Ineffective 1 2 mg dexamethasone, 20 mg triamcinolone with 2 a. Nortriptyline 25 mg a. Ineffective 7 3 mg dexamethasone with 3 a. Ibuprofen 600 mg, as needed a. Partial 1 3 mg dexamethasone with 4 a. Nortriptyline 75 mg a. Ineffective 1 3 mg dexamethasone with with epinephrine 5 None NA 3 2 mg dexamethasone, 20 mg triamcinolone with <1 week 90% reduction in pain intensity <1 week 50% reduction in pain intensity lasting 4 weeks 62 Remission 8 Improved, recurrent trochleitis NR Complete relief 39 Remission Immediate Partial relief for only 12 h NR Complete relief, durability of 7 months, then 12 months, and then ongoing relief 7 Chronic trochlear 47 Remission 6 None NA 1 4 mg dexamethasone NR Complete relief 10 Remission 7 a. OxyContin 30 mg twice daily 8 a. Amitriptyline 50 mg b. Gabapentin 1800 mg c. Intravenous methylprednisolone 500 mg d. Dilaudid, as needed e. Methotrexate f. Azathioprine g. Infliximab a. Partial 12 2 mg dexamethasone, 40 mg triamcinolone with 0.25 ml 2% lidocaine a. Ineffective b. Ineffective c. Ineffective d. Ineffective e. Ineffective f. Ineffective g. Complete 8 3 mg dexamethasone, 40 mg triamcinolone with 9 None NA 1 3 mg dexamethasone, ml 2% lidocaine 10 a. Nortriptyline 50 a. Improved migraines, but not trochleitis pain 1 3 mg dexamethasone, 0.25 ml 2% lidocaine 2 4 days Injections 1 8, 10: partial relief, average durability of 22 days Injections 9, 11 12: ineffective Immediate Partial relief, durability range 2 5 days 27 Improved, persistent trochleitis 42 Remission Immediate Partial None Unknown Immediate Complete 25 Remission

7 Prognosis of trochlear s 7 Table 2 (Continued ) no. Prior treatments other than trochlear injection Efficacy of treatments other than trochlear Total number of trochlear Injected material (total volume 1 ml for all cases) 11 a. Propranolol 120 mg b. Gabapentin 1800 mg c. Prednisone 60 mg d. Azathioprine e. Rituximab f. External beam radiation (2000 cgy in 10 fractions) 12 a. Indomethacin 50 mg three times daily b. Acetazolamide 250 mg twice daily a. Ineffective b. Ineffective c. Partial d. Partial e. Partial f. Complete a. Partial b. Partial 5 3 mg dexamethasone, 18 3 mg dexamethasone, 13 a. Prednisone b. Azathioprine a. Complete b. Partial 3 2 mg dexamethasone, 0.1 ml 2% lidocaine 14 None NA 5 3 mg dexamethasone, 15 a. Amitriptyline b. Carbamazepine a, b. Unable to tolerate either 3 3 mg dexamethasone, and with epinephrine 16 None NA 4 2 mg dexamethasone, 20 mg triamcinolone 17 None NA 2 2 mg dexamethasone, 20 mg triamcinolone 0.25 ml 2% lidocaine 18 None NA 1 3 mg dexamethasone, 19 a. Topiramate 100 mg b. Amitriptyline 40 mg c. Prednisone 40 mg a. Ineffective b. Ineffective c. Partial ml 2% lidocaine 2 3 mg dexamethasone, 0.25 ml 2% lidocaine Onset of action of Outcome of Time from diagnosis to (months) Outcome at Unknown 1 2, 4 5: partial, durability range 6 weeks to 2 months 3: Ineffective 30 Remission Unknown 1 3, 5 7, 9 1: partial, durability range 2 6 weeks, except last two lasted 5, then 7 months 4, 8: ineffective Unknown Complete, durability 9 12 months 32 Improved, recurrent trochlear 68 Improved, recurrent trochlear Unknown Complete, durability of 3 4 months 20 Improved, recurrent trochleitis NA Ineffective 34 Chronic trochlear Unknown 1 3: partial, durability range of 8 12 months 4: ineffective Unknown 1, 2: partial, durability of 2 months each 44 Persistent trochleitis 6 Improved, recurrent trochlear Unknown Complete 37 Remission 10 days 1, 2: complete (right eye), partial (left eye) 1 Improved, low-grade (2/10) trochlear on left side, remission on right side

8 8 J. H. Smith, J. A. Garrity and C. J. Boes Table 2 (Continued ) no. Prior treatments other than trochlear injection Efficacy of treatments other than trochlear Total number of trochlear Injected material (total volume 1 ml for all cases) 20 a. Oxycodone 15 mg b. Methotrexate 0.5 ml per week c. Cyclophosphamide d. Rituximab e. Prednisone 60 mg a. Partial b. Ineffective c. Partial d. Partial e. Partial 2 1 mg dexamethasone, 0.1 ml 2% lidocaine 21 None NA 3 3 mg dexamethasone, 0.1 ml 2% lidocaine 22 a. Indomethacin 75 mg three times daily b. Amitriptyline 50 mg c. Gabapentin 1800 mg d. Propranolol 80 mg e. Topiramate 50 mg f. Botulinum Toxin A, 150 units 9 2 g. Prednisone 80 mg h. Methotrexate i. External beam radiation (25 CGy over 14 fractions) a f, i. Ineffective g. Partial h. Near-complete 6 3 mg dexamethasone, 20 mg triamcinolone, 23 None NA 4 1 3: 2 mg dexamethasone, 20 mg triamcinolone and 0.25 ml 2% lidocaine with epinephrine 4: 4 mg dexamethasone, 20 mg triamcinolone, with epinephrine 24 a. Nortriptyline 50 mg b. Prednisone 60 mg a. Partial b. Partial 1 3 mg dexamethasone, 25 None NA 7 3 mg dexamethasone, 0.2 ml 2% lidocaine NA, not applicable. Onset of action of Outcome of Time from diagnosis to (months) Outcome at Unknown 1: ineffective 2: partial 49 Chronic trochlear Unknown 1, 2: ineffective 3: complete 16 Remission Unknown Partial, durability 2 3 weeks 64 Improved, low-grade (1-2/10) persistent trochlear <7 days Complete, durability of 1 2: 3 months, 3: 6 months, 4: remission 57 Remission NA Ineffective 40 Chronic trochlear <7 days Partial, average durability of 6 weeks 54 Recurrent trochlear

9 Prognosis of trochlear s 9 Disclosure of conflicts of interest The authors declare no financial or other conflicts of interest. References 1. Helveston EM, Merriam WW, Ellis FD, Shellhamer RH, Gosling CG. The trochlea. A study of the anatomy and physiology. Ophthalmology 1982; 89: Pareja JA, Sanchez del Rio M. Primary trochlear and other trochlear painful disorders. Curr Pain Headache Rep 2006; 10: Yanguela J, Sanchez-del-Rio M, Bueno A, et al. Primary trochlear : a new cephalgia generated and modulated on the trochlear region. Neurology 2004; 62: Tychsen L, Tse DT, Ossoinig K, Anderson RL. Trochleitis with superior oblique myositis. Ophthalmology 1984; 91: Yanguela J, Pareja JA, Lopez N, Sanchez Del Rio M. Trochleitis and migraine. Neurology 2002; 58: The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: Robbins MS, Grosberg BM, Napchan U, Crystal SC, Lipton RB. Clinical and prognostic subforms of new daily-persistent. Neurology 2010; 74: Pareja JA, Cuadrado ML, Porta-Etessam J, et al. Idiopathic ophthalmodynia and idiopathic rhinalgia: two topographic facial pain syndromes. Headache 2010; 50: