European Journal of Neurology 2014, 21: 577–585
doi:10.1111/ene.12312
Clinical features and long-term prognosis of trochlear headaches J. H. Smitha, J. A. Garrityb and C. J. Boesc a
Department of Neurology, University of Kentucky, Lexington, KY; bDepartment of Ophthalmology, Mayo Clinic, Rochester, MN; and
c
Department of Neurology, Mayo Clinic, Rochester, MN, USA
Keywords:
chronic daily headache, migraine, neuroophthalmology, ocular movements, secondary headache disorders, trochlea, trochleitis Received 13 September 2013 Accepted 21 October 2013
Background and purpose: Trochlear headaches are a recently recognized cause of headache, 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 headache seen at the Mayo Clinic between 10 July 2007 and 28 June 2012 were identified. Results: The diagnosis of trochlear headache was not recognized by the referring neurologist or ophthalmologist in any case. Patients most often presented with a new daily from onset headache (n = 22, 88%). The most characteristic headache 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 injections 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 headaches are poorly recognized, have characteristic clinical features, and often require serial injections 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 headache (PTRH), has also been described [3]. Treatment of all trochlear headaches 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.: 859-323-5661; fax: 859-323-5943; e-mail: jonathan.smith@uky. edu).
© 2013 The Author(s) European Journal of Neurology © 2013 EFNS
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 headache 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 headache could be definitively assigned and adequate clinical documentation was available. In our paper, the general term trochlear headache is used to describe any headache referable to the troch-
577
578
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 headache which then triggers a secondary migraine attack [2,5]. The diagnostic criteria for headache 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 headache 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 headache 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 headache 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 headache (n = 5), no diagnosis (n = 4), hemicrania continua (n = 2) and atypical facial pain (n = 1). All patients with a prior headache 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 headache began immediately following orbital surgery for removal of an optic nerve
sheath meningioma. One individual developed the headache 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 headaches 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 headache 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 headache was most often focused at the medial eyebrow, 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 headache 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 headache syndrome. These headaches 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 © 2013 The Author(s) European Journal of Neurology © 2013 EFNS
55/F 29/M 77/F
34/F 32/M
57/F
38/F
36/F
63/M
18/F
56/F 42/F
71/F
65/F 57/F 40/F
74/F
58/F
46/F
23/F 48/M
4 5
6
7
8
9
10
11 12
13
14 15 16
17
18
19
20 21
Age/gender
1 2 3
Case no.
© 2013 The Author(s) European Journal of Neurology © 2013 EFNS
Trochleitis, GPA/L Trochleitis/L
PTRH, TM/Bil
PTRH/Bil
PTRH/Bil
Trochleitis/Bil PTRH/R Trochleitis/R
PTRH/Bil
Trochleitis, GPA/L PTRH/R
Trochleitis, TM/L
Trochleitis, incomplete Behcet’s/Bil Trochleitis/ Tolosa Hunt/Bil Trochleitis/R
PTRH/Bil
PTRH, TM/Bil PTRH/Bil
PTRH, TM/R Trochleitis, TM/Bil PTRH/R
Diagnosis/laterality
Acute Acute
Acute
Acute
Acute
Acute Acute Acute
Acute
Acute Acute
Acute
Acute
Acute
Acute
Acute
Acute Acute
Acute Acute Acute
Acuity of onset
Continuous Continuous
Continuous
Continuous
Continuous
Continuous Continuous Continuous
Continuous
Continuous Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous Continuous
Continuous Continuous Continuous
Continuous or episodic
Table 1 Clinical features of patients with trochlear headache
Eyebrow Orbit
Retroorbital Retroorbital Medial forehead Medial eyebrow Medial forehead Medial eyebrow
Medial eyebrow Eyebrow Medial eyebrow Periorbital
Periorbital
Periorbital
Orbit
‘Eye’
Orbit Orbit, medial Medial eyebrow Periorbital Retroorbital
Location
Forehead, bridge of nose Orbit None
Forehead, temple None
Eyebrow None Retro-orbital
None
None Forehead
None
None
Forehead
None
NR
Holocephalic Temple
None None None
Radiation
Throb/7 Pressure/2
Pressure/4
Ache
Pressure/6
Pressure, sharp/4 Pressure/4 Pressure/4
Dull/3
Ache/2 Pressure/5
NR
Ache/8
Pressure/7
Ache/7
NR/7
Throb Pressure/6
Pressure/6 Throb/6 Ache
Quality/ intensity
PHT, DIP None
PHT
PHT, PHN
PHT
PHT, DIP PHT PHT, DIP
PHT
PHT, DIP NR
NR
NR
NA, DIP
NA, DIP
DIP
NA, PHT, DIP NR
NA, PHT NA, PHT, DIP NR
Associated symptoms
EM EM
EM
Touch
EM
EM EM EM
EM
EM EM
NR
None
EM
EM
EM
None EM
‘Reading’ EM None
Aggravating factors
None None
None
None
Warmth
None Warmth None
Warmth
None None
NR
None
Eye patch
None
NR
None NR
None None None
Relieving factors
GPA ETTH
Depression
EMO, depression
History of gastric bypass
GAD
History of encephalitis in childhood, unknown cause EMO
CMA EMO Left optic nerve meningioma EMO, IED Right orbital dermoid Graves ophthalmopathy s/p decompression 1 month prior to onset CMO
Relevant comorbidity
Prognosis of trochlear headaches
579
580
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 headache; 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.
None EM Dull/7 None Retroorbital Continuous Subacute 39/F 25
PTRH/Bil
Continuous 41/F 24
Trochleitis, orbital MALT lymphoma/L
Subacute
Periorbital
Hemicranial/neck
Throbbing/7
PHT, blurry vision None
None
None
EMO GBM, s/p WBRT (1 year before onset) +Radiationinduced retinopathy and optic neuropathies History of anticardiolipin antibody None None Eye closure EM None PHT, DIP PHT Pressure Pressure/6 Eyebrow Forehead Periorbital Medial eyebrow Continuous Continuous Subacute Acute 35/F 48/M 22 23
Trochleitis, TM/L PTRH/L
Continuous or episodic Age/gender Case no.
Table 1 (Continued )
Diagnosis/laterality
Acuity of onset
Location
Radiation
Quality/ intensity
Associated symptoms
Aggravating factors
Relieving factors
Relevant comorbidity
J. H. Smith, J. A. Garrity and C. J. Boes
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 headache 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 injections 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 injections per patient was three (range 1–12). Amongst the 17 individuals receiving repeat injections 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 injections. Of patients with an initially ineffective first injection, subsequent injections 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 headache practice, trochlear headaches 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 © 2013 The Author(s) European Journal of Neurology © 2013 EFNS
Prognosis of trochlear headaches
(a)
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)
presentation could be easily confused with new-daily persistent headache, which is generally thought of as extremely difficult to treat [7]. However, the prevalence of undiagnosed trochlear headache in patients with presumed new-daily persistent headache 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 injections 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 injections, 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, © 2013 The Author(s) European Journal of Neurology © 2013 EFNS
581
(b)
(d)
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 headache 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, headache syndrome.
a. Ibuprofen 600 mg, as needed
a. Nortriptyline 75 mg
None
None a. OxyContin 30 mg twice daily
a. Amitriptyline 50 mg b. Gabapentin 1800 mg c. Intravenous methylprednisolone 500 mg d. Dilaudid, as needed e. Methotrexate f. Azathioprine g. Infliximab None
a. Nortriptyline 50
3
4
5
6 7
8
10
9
2
a. Indomethacin 50 mg TID b. Botulinum toxin A injections, 150 units 9 1 c. Melatonin 3 mg a. Nortriptyline 25 mg
1
Case no.
Prior treatments other than trochlear injection
a. Improved migraines, but not trochleitis pain
NA
a. Ineffective b. Ineffective c. Ineffective d. Ineffective e. Ineffective f. Ineffective g. Complete
NA a. Partial
NA
a. Ineffective
a. Partial
a. Ineffective
a. Partial b. Partial c. Ineffective
Efficacy of treatments other than trochlear injections
1
1
8
1 12
3
1
1
7
1
Total number of trochlear injections
Table 2 Treatment outcomes of patients with trochlear headache
3 mg dexamethasone, 0.125 ml 2% lidocaine with epinephrine 3 mg dexamethasone, 0.25 ml 2% lidocaine with epinephrine
3 mg dexamethasone, 40 mg triamcinolone with 0.25 ml 2% lidocaine
4 mg dexamethasone 2 mg dexamethasone, 40 mg triamcinolone with 0.25 ml 2% lidocaine
3 mg dexamethasone with 0.25 ml 2% lidocaine with epinephrine 3 mg dexamethasone with 0.25 ml 2% lidocaine with epinephrine 3 mg dexamethasone with 0.25 ml 2% lidocaine with epinephrine 2 mg dexamethasone, 20 mg triamcinolone with 0.25 ml 2% lidocaine
2 mg dexamethasone, 20 mg triamcinolone with 0.25 ml 2% lidocaine
Injected material (total volume 1 ml for all cases)
Immediate
Immediate
Immediate
NR 2–4 days
NR
Immediate
Complete
Partial
Complete relief, durability of 7 months, then 12 months, and then ongoing relief Complete relief Injections 1–8, 10: partial relief, average durability of 22 days Injections 9, 11–12: ineffective Partial relief, durability range 2–5 days
Partial relief for only 12 h
50% reduction in pain intensity lasting 4 weeks Complete relief
doi:10.1111/ene.12312
Clinical features and long-term prognosis of trochlear headaches J. H. Smitha, J. A. Garrityb and C. J. Boesc a
Department of Neurology, University of Kentucky, Lexington, KY; bDepartment of Ophthalmology, Mayo Clinic, Rochester, MN; and
c
Department of Neurology, Mayo Clinic, Rochester, MN, USA
Keywords:
chronic daily headache, migraine, neuroophthalmology, ocular movements, secondary headache disorders, trochlea, trochleitis Received 13 September 2013 Accepted 21 October 2013
Background and purpose: Trochlear headaches are a recently recognized cause of headache, 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 headache seen at the Mayo Clinic between 10 July 2007 and 28 June 2012 were identified. Results: The diagnosis of trochlear headache was not recognized by the referring neurologist or ophthalmologist in any case. Patients most often presented with a new daily from onset headache (n = 22, 88%). The most characteristic headache 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 injections 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 headaches are poorly recognized, have characteristic clinical features, and often require serial injections 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 headache (PTRH), has also been described [3]. Treatment of all trochlear headaches 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.: 859-323-5661; fax: 859-323-5943; e-mail: jonathan.smith@uky. edu).
© 2013 The Author(s) European Journal of Neurology © 2013 EFNS
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 headache 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 headache could be definitively assigned and adequate clinical documentation was available. In our paper, the general term trochlear headache is used to describe any headache referable to the troch-
577
578
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 headache which then triggers a secondary migraine attack [2,5]. The diagnostic criteria for headache 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 headache 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 headache 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 headache 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 headache (n = 5), no diagnosis (n = 4), hemicrania continua (n = 2) and atypical facial pain (n = 1). All patients with a prior headache 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 headache began immediately following orbital surgery for removal of an optic nerve
sheath meningioma. One individual developed the headache 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 headaches 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 headache 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 headache was most often focused at the medial eyebrow, 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 headache 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 headache syndrome. These headaches 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 © 2013 The Author(s) European Journal of Neurology © 2013 EFNS
55/F 29/M 77/F
34/F 32/M
57/F
38/F
36/F
63/M
18/F
56/F 42/F
71/F
65/F 57/F 40/F
74/F
58/F
46/F
23/F 48/M
4 5
6
7
8
9
10
11 12
13
14 15 16
17
18
19
20 21
Age/gender
1 2 3
Case no.
© 2013 The Author(s) European Journal of Neurology © 2013 EFNS
Trochleitis, GPA/L Trochleitis/L
PTRH, TM/Bil
PTRH/Bil
PTRH/Bil
Trochleitis/Bil PTRH/R Trochleitis/R
PTRH/Bil
Trochleitis, GPA/L PTRH/R
Trochleitis, TM/L
Trochleitis, incomplete Behcet’s/Bil Trochleitis/ Tolosa Hunt/Bil Trochleitis/R
PTRH/Bil
PTRH, TM/Bil PTRH/Bil
PTRH, TM/R Trochleitis, TM/Bil PTRH/R
Diagnosis/laterality
Acute Acute
Acute
Acute
Acute
Acute Acute Acute
Acute
Acute Acute
Acute
Acute
Acute
Acute
Acute
Acute Acute
Acute Acute Acute
Acuity of onset
Continuous Continuous
Continuous
Continuous
Continuous
Continuous Continuous Continuous
Continuous
Continuous Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous Continuous
Continuous Continuous Continuous
Continuous or episodic
Table 1 Clinical features of patients with trochlear headache
Eyebrow Orbit
Retroorbital Retroorbital Medial forehead Medial eyebrow Medial forehead Medial eyebrow
Medial eyebrow Eyebrow Medial eyebrow Periorbital
Periorbital
Periorbital
Orbit
‘Eye’
Orbit Orbit, medial Medial eyebrow Periorbital Retroorbital
Location
Forehead, bridge of nose Orbit None
Forehead, temple None
Eyebrow None Retro-orbital
None
None Forehead
None
None
Forehead
None
NR
Holocephalic Temple
None None None
Radiation
Throb/7 Pressure/2
Pressure/4
Ache
Pressure/6
Pressure, sharp/4 Pressure/4 Pressure/4
Dull/3
Ache/2 Pressure/5
NR
Ache/8
Pressure/7
Ache/7
NR/7
Throb Pressure/6
Pressure/6 Throb/6 Ache
Quality/ intensity
PHT, DIP None
PHT
PHT, PHN
PHT
PHT, DIP PHT PHT, DIP
PHT
PHT, DIP NR
NR
NR
NA, DIP
NA, DIP
DIP
NA, PHT, DIP NR
NA, PHT NA, PHT, DIP NR
Associated symptoms
EM EM
EM
Touch
EM
EM EM EM
EM
EM EM
NR
None
EM
EM
EM
None EM
‘Reading’ EM None
Aggravating factors
None None
None
None
Warmth
None Warmth None
Warmth
None None
NR
None
Eye patch
None
NR
None NR
None None None
Relieving factors
GPA ETTH
Depression
EMO, depression
History of gastric bypass
GAD
History of encephalitis in childhood, unknown cause EMO
CMA EMO Left optic nerve meningioma EMO, IED Right orbital dermoid Graves ophthalmopathy s/p decompression 1 month prior to onset CMO
Relevant comorbidity
Prognosis of trochlear headaches
579
580
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 headache; 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.
None EM Dull/7 None Retroorbital Continuous Subacute 39/F 25
PTRH/Bil
Continuous 41/F 24
Trochleitis, orbital MALT lymphoma/L
Subacute
Periorbital
Hemicranial/neck
Throbbing/7
PHT, blurry vision None
None
None
EMO GBM, s/p WBRT (1 year before onset) +Radiationinduced retinopathy and optic neuropathies History of anticardiolipin antibody None None Eye closure EM None PHT, DIP PHT Pressure Pressure/6 Eyebrow Forehead Periorbital Medial eyebrow Continuous Continuous Subacute Acute 35/F 48/M 22 23
Trochleitis, TM/L PTRH/L
Continuous or episodic Age/gender Case no.
Table 1 (Continued )
Diagnosis/laterality
Acuity of onset
Location
Radiation
Quality/ intensity
Associated symptoms
Aggravating factors
Relieving factors
Relevant comorbidity
J. H. Smith, J. A. Garrity and C. J. Boes
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 headache 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 injections 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 injections per patient was three (range 1–12). Amongst the 17 individuals receiving repeat injections 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 injections. Of patients with an initially ineffective first injection, subsequent injections 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 headache practice, trochlear headaches 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 © 2013 The Author(s) European Journal of Neurology © 2013 EFNS
Prognosis of trochlear headaches
(a)
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)
presentation could be easily confused with new-daily persistent headache, which is generally thought of as extremely difficult to treat [7]. However, the prevalence of undiagnosed trochlear headache in patients with presumed new-daily persistent headache 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 injections 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 injections, 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, © 2013 The Author(s) European Journal of Neurology © 2013 EFNS
581
(b)
(d)
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 headache 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, headache syndrome.
a. Ibuprofen 600 mg, as needed
a. Nortriptyline 75 mg
None
None a. OxyContin 30 mg twice daily
a. Amitriptyline 50 mg b. Gabapentin 1800 mg c. Intravenous methylprednisolone 500 mg d. Dilaudid, as needed e. Methotrexate f. Azathioprine g. Infliximab None
a. Nortriptyline 50
3
4
5
6 7
8
10
9
2
a. Indomethacin 50 mg TID b. Botulinum toxin A injections, 150 units 9 1 c. Melatonin 3 mg a. Nortriptyline 25 mg
1
Case no.
Prior treatments other than trochlear injection
a. Improved migraines, but not trochleitis pain
NA
a. Ineffective b. Ineffective c. Ineffective d. Ineffective e. Ineffective f. Ineffective g. Complete
NA a. Partial
NA
a. Ineffective
a. Partial
a. Ineffective
a. Partial b. Partial c. Ineffective
Efficacy of treatments other than trochlear injections
1
1
8
1 12
3
1
1
7
1
Total number of trochlear injections
Table 2 Treatment outcomes of patients with trochlear headache
3 mg dexamethasone, 0.125 ml 2% lidocaine with epinephrine 3 mg dexamethasone, 0.25 ml 2% lidocaine with epinephrine
3 mg dexamethasone, 40 mg triamcinolone with 0.25 ml 2% lidocaine
4 mg dexamethasone 2 mg dexamethasone, 40 mg triamcinolone with 0.25 ml 2% lidocaine
3 mg dexamethasone with 0.25 ml 2% lidocaine with epinephrine 3 mg dexamethasone with 0.25 ml 2% lidocaine with epinephrine 3 mg dexamethasone with 0.25 ml 2% lidocaine with epinephrine 2 mg dexamethasone, 20 mg triamcinolone with 0.25 ml 2% lidocaine
2 mg dexamethasone, 20 mg triamcinolone with 0.25 ml 2% lidocaine
Injected material (total volume 1 ml for all cases)
Immediate
Immediate
Immediate
NR 2–4 days
NR
Immediate
Complete
Partial
Complete relief, durability of 7 months, then 12 months, and then ongoing relief Complete relief Injections 1–8, 10: partial relief, average durability of 22 days Injections 9, 11–12: ineffective Partial relief, durability range 2–5 days
Partial relief for only 12 h
50% reduction in pain intensity lasting 4 weeks Complete relief
