Head Trauma and Cluster Headache Gian Camillo Manzoni, MD, Giorgio Lambru, MD, and Paola Torelli, MD

Corresponding author Gian Camillo Manzoni, MD Headache Centre, Department of Neurology, University of Parma and University Centre for Adaptive Disorders and Headache, Via Gramsci, 14, 43100 Parma, Italy. E-mail: [email protected] Current Pain and Headache Reports 2006, 10:130 –136 Current Science Inc. ISSN 1531-3433 Copyright © 2006 by Current Science Inc.

Post-traumatic cluster headache (CH) is a very rare occurrence. Only one such case has been verified in the literature thus far. On the other hand, a non-casual association is almost certain to exist between previous head trauma and future CH onset. Considering the generally long interval of time separating the two events, it is not easy to accurately explain the nature of the head trauma/ CH relationship. Head trauma may damage extra- or intracranial peripheral or central nervous structures, hence predisposing to the future development of CH. However, there also is the possibility that previous head injuries may be more frequent among CH patients because of their lifestyle, which may leave them more exposed to the risk of traumatic events.

Introduction Cluster headache (CH) is a primary headache characterized by highly specific clinical features. The name itself, which was first termed for this headache in 1952 by Kunkle et al. [1], underlines one of its most distinctive traits—a typical temporal pattern that recurs periodically. The most diagnostically significant clinical characteristics of CH are strictly unilateral pain location and severe pain intensity, the presence of accompanying ipsilateral autonomic symptoms, and the restless behavior of patients during attacks [2,3•]. Its prevalence rate being less than 1% of the general population [4], epidemiologists typically consider CH as a rare disorder that affects men more than women [5]. However, recent studies have demonstrated that it is spreading among the general population at a greater rate than was thought in the past (~ 3%) [6,7], suggesting a progressive reduction over the years of its typical male predominance [8]. Despite recent evidence indicating that CH could be considered a primary neurovascular disorder [9] and that

the hypothalamus may play a crucial role in its pathogenetic mechanisms [10], the underlying cause of this type of headache remains unknown. Two schools of thought are currently prevailing among researchers; one is based on epidemiologic data of familial occurrence and stresses the role of a possible genetic predisposition [11,12], and the other points to specific features that have been known for a long time to be shared by most CH sufferers (eg, the tendency to overindulge in non-essential habits such as smoking) and perceives patients’ lifestyle as a factor of foremost importance [13••,14]. Obviously, these two lines of reasoning are not in conflict and are not mutually exclusive, possibly representing the two sides of the same coin. Although CH should be considered in every respect as a type of primary headache, several reports have been published in the past few years about patients with headache forms that were totally comparable with CH and were strictly associated with organic disorders affecting the head (eg, vascular disease, infections, inflammation, tumors, glaucoma, and dental disease) [15]. In addition to these reports regarding what actually could be secondary CH forms, a few studies have pointed to a possibly significant presence of recent or remote head trauma in the past medical histories of CH patients. The subject is extremely interesting, both clinically and pathogenetically, and is certainly worth discussing in its various aspects, even though our present-day knowledge is still too scarce to explain the true nature of what likely is a complex relationship.

Literature Review In 1949, Hunter and Mayfield [16] reported eight cases in which the upper cervical nerves were injured following a whiplash injury to the neck, which, in addition to discomfort in the suboccipital region, produced attacks closely resembling those of CH. However, the authors did not provide any accurate information about the interval of time separating head trauma and the onset of headache symptoms, or about the exact clinical features of headache. In 1956, Symonds [17] presented a case series of 17 patients with a particular type of headache that fully matched CH characteristics. Two of these patients (11.8%) had significant head injuries in their medical histories. The latency between head trauma and onset of headache was 5 years in one case and 25 years in the other.

Head Trauma and Cluster Headache Manzoni et al. 131

In 1958, Friedman and Mikropoulos [18] wrote about “a past medical history of some type of head injury, with or without loss of consciousness” in eight of their 50 patients with CH (16%), without providing any information about the latency between the two events. In 1971, Lance and Anthony [19], after reviewing the clinical features of 60 of their CH patients, reported that eight (13.3%) had suffered a head trauma and in four of them, the site of the injury could conceivably be relevant to the ensuing CH. The authors provided a detailed description of these four cases. One patient had required extensive plastic surgery for facial and scalp lacerations following a car accident, and left-sided CH started 21 months after injury. The second had gravel embedded in the right temple from a road accident at the age of 17 years, and CH involving the right temple and right side of the face began 7 years later. The third patient developed right frontotemporal CH 30 years after a shotgun injury to the forehead, in which fragmented pellets were still embedded. The fourth patient experienced a blow to the forehead at the age of 22 years, following which he was subject to a dull right-sided headache for 3 years. Eight years after the injury, he developed right-sided CH involving the right forehead and right side of the face. In 1979, in an interesting “Headache Round” aimed at discussing the possible role of eye structures in the pathophysiology of CH, Rogado and Graham [20] described five patients who suffered from CH on the side of an absent eye. In one patient, the eye had been removed after CH onset in the ridiculous attempt to “relieve the headaches in his eye,” and for another patient, there was no information available about the latency between eye loss and CH onset. However, in the remaining three patients, CH had clearly started after the ipsilateral exenteration, with latencies of 3, 4, and approximately 15 years. In all three patients, the bulb was removed after traumatic injury and it is interesting to note that in two of the three patients, other head traumas had occurred before CH onset. The first of these cases was a 47-year-old engineer. While he was playing as a child with a cousin who had just watched a film featuring American Indians and cowboys, an arrow struck his right eye, which had to be removed. Shortly after turning 20, he began to suffer from right-sided CH. The second case was a 26-year-old carpenter. At the age of 16 years, a large steel chip got into his right eye and penetrated it like the point of a screw driver. The injured eye was spared for approximately 1 year. At the age of 17 years, he was violently hit again in the same eye by a fist blow dealt by someone who had mistaken him for a rival. The eye bled and had to be enucleated. At 21 years of age, perhaps because of the artificial eye, he walked into a plank that sprang and hit him on the right side of his head. Right-sided CH started shortly thereafter. The third case was a 49-year-old librarian. In his youth, he had been an accomplished athlete. At the age

of 18 years, a basketball with an odd spin caught his open right eye and tore the retina. The eye damage was underestimated for years, despite progressive vision loss and persistent eye inflammation. Then, when he was 41 years of age, the eye had to be enucleated because of a secondary glaucoma. At 44 years of age, following a car accident, he hit his head against the rear-view mirror and got a cutting wound in the middle of his forehead. Four months later, he began to suffer from right-sided CH. In 1980, Kudrow [5] reviewed a case series of 230 male CH patients and found a history of significant head injury in 5.2% of them (ie, a lower figure than previously reported by other authors) [17–19]. In 1983, Manzoni et al [21] reported a case series of 180 CH sufferers (158 men and 22 women), including 161 with episodic CH and 19 with chronic CH. In 41 cases (22.8%), CH onset was preceded by head trauma. In 20 of these cases (11.1%), patients had suffered head injuries with loss of consciousness. This kind of head trauma was significantly more frequent in CH patients than in two control populations of migraine patients and tension-type headache patients, respectively. In all of the cases in whom head injury with loss of consciousness seemed to have been unilateral (12 of 20 patients), the patients reported that the side involved was the same where CH would occur later. Mean latency between head trauma and CH onset was 14.4 years; if only unilateral injuries are considered, it was 9.2 years. Head injuries with loss of consciousness were more frequent in patients with chronic CH (21.1%) than in those with episodic CH (9.9%). In the same case series [21], another 15 CH patients had undergone some type of facial surgical procedure before the onset of CH. In 11 of them (6.1%), the side operated on was the same as that of CH, and the latency between the two events averaged 5 years. Between 1983 and 1987, three more cases of CH were reported after exenteration of the ipsilateral orbit. In two of them [22,23], the surgical procedure was needed to remove a tumor—a squamous cell carcinoma in one patient (who developed CH 1 year later) [22] and a sclerosing basal-cell carcinoma in the other patient [23]. However, in the latter case, the description of the headache that the patient developed 6 months after surgery seems to cast some doubts on the reliability of the CH diagnosis. The third case [24] was a man who at the age of 34 years, had his right eye removed because of a blast injury, and at 52 years of age, began to suffer from CH on the same side. In 1987, Reik [25] described four patients who developed CH following head trauma. The first patient was a 45-year-old man who had never experienced CH until the age of 42 years, which is when he fell and suffered a head injury on the right side of his forehead and experienced loss of consciousness for approximately 15 minutes. When he regained consciousness, he initially complained of daily bifrontal headache for several weeks,

132 Cluster Headache

then developed a right-sided frontal headache that spread from the injury site to the left temple. The attacks could vary in intensity, but continued daily for approximately 3 years. Finally, this form of headache was replaced by typical left-sided CH (ie, contralateral to the side of the head trauma). The second patient was a 39-year-old man who was headache-free until the age of 35 years when he was hit in the forehead by a cable and lost consciousness for several minutes. When he regained consciousness, he began to complain of pain in the left supraorbital region radiating to the occipital region. This moderate background pain persisted continuously in the following weeks, being occasionally overlapped by a more intense pain in the same site for 1 to 2 hours. During the following months, the background pain disappeared while the occasional pain exacerbations evolved to typical left-sided CH, with attacks occurring once or more than once daily. Until observation, the patient was never headache-free for more than a few days. Therefore, this was a case of leftsided chronic CH occurring very shortly after a frontal head trauma on an unspecified side (or perhaps in the middle of the forehead). The third patient was a 24-year-old man. At the age of 22 years, he struck his left temple and forehead during a motor vehicle accident without loss of consciousness. One month later, he began to suffer from ipsilateral CH, with clusters lasting for 1 month. After a 5-month interval, the CH attacks recurred with no more symptom-free intervals until the time of observation 18 months later. Therefore, this was a case of chronic CH ipsilateral to head trauma. The fourth patient was a 63-year-old man who had suffered a concussion and frontal skull fracture during a bomb blast at the age of 20 years. After a few weeks, he developed chronic left-sided CH, which still persists to this day after 43 years, except for a headache-free interval of a few months approximately 20 years after onset. Only the third of the four patients described by Reik [25] presented with true side correspondence between head trauma and CH. At the Thirtieth Annual Meeting of the American Association for the Study of Headache (AASH) held in San Francisco, California, on June 18–19, 1988, Mathew and Rueveni [26] presented a personal case series (unfortunately, only one abstract is available as documentation) involving patients who began to suffer from a type of headache that was similar to CH, but with certain differences. The authors reported that blunt head trauma, fractures of the facial bones, trauma in and around the eyes, and surgical treatment for other neurosurgical conditions triggered CH-like attacks. The side of the headache corresponded with the side of the maximum trauma or surgical structural disruption. The time pattern of the headache was chronic. Unfortunately, more accurate clinical information is

unavailable. Mathew himself, during a platform presentation on January 25, 1992 at the AASH Comprehensive Course in Mechanisms and Management of Headache in Scottsdale, Arizona, said that his case series included 11 patients, but he did not state exactly the total number of CH patients who were observed. In 1990, Formisano et al [27] described a 39-year-old patient who developed a right-sided cluster-like headache 45 days after a head trauma in the right frontotemporal region, with subsequent chronic subdural hematoma. In 1992, Turkewitz et al. [28•] reported the case of a 31-year-old woman who developed a typical form of CH on the right side 6 days after suffering a “minor head trauma with no loss of consciousness in an auto accident.” The authors provided an accurate review of the existing literature, but unfortunately failed to give more detailed indications about the site of the trauma suffered by the patient. In a case-control study on the epidemiology of CH [29], which was conducted in 1995 at nine headache centers in Italy, previous head trauma was found in 37 of 120 (30.8%) CH patients versus 19 of 120 (15.8%) control subjects, the odds ratio being 2.50 (95% CI, 1.28–4.88). In 15 of 120 (12.5%) CH patients, head injuries were accompanied by loss of consciousness. In 1997, Evers et al. [30] reported another case of CH after orbital exenteration. The patient was a 37-year-old man who had lost eyesight in his right eye after an accident occurring at work at the age of 27 years. A piece of wire perforated his right bulb and damaged the retina and the optical nerve. The wire was removed and the wound was closed microsurgically on the same day. In the next 2 months, the lens and the vitreous body were removed and replaced by silicon. The right bulb was finally enucleated 7 months after the injury due to secondary glaucoma and to its shrinkage. Approximately 3 weeks after exenteration of the orbit, the patient developed an episodic form of ipsilateral CH, which recurred in one to two clusters a year until the time of observation. In 1999, Manzoni [13••] tried to gather as much detailed data as possible on any previous head injuries (most notably, on their extent, site, type, and date of occurrence) in 374 male CH patients (306 with episodic CH, 42 with chronic CH, and 26 with CH periodicity undetermined) and in 374 age- and gender-matched control subjects with migraine. A previous head trauma was present in 138 CH patients (36.9%) compared with 63 control subjects (16.9%; (P < 0.001). In 50 CH patients (13.4%), the head injuries were accompanied by loss of consciousness. Head injuries, with and without loss of consciousness, appeared more frequent in the medical histories of patients with chronic CH (19.0% and 35.7%, respectively) than of those with episodic CH (13.4% and 23.5%, respectively). Among chronic CH patients, those who reported a previous head injury more frequently with or without loss of consciousness

Head Trauma and Cluster Headache Manzoni et al. 133

Table 1. Head injury with or without loss of consciousness in patients with cluster headache: time relation between occurrence of injury and onset of cluster headache Head injury before CH onset Head injury with LOC

Head injury after CH onset

Number

Percentage

Average latency, y

Number

Percentage

Total number

40

90.9

12

4

9.1

44

Side +*

16

84.2

11.1

3

15.8

19

Side -†

11

100

12.7

0

NA

11

Side ?

13

92.9

12.4

1

7.1

14

59

71.1

8.9

24

28.9

83

Side +*

23

85.2

9.1

4

14.8

27

Side -

27

64.3

9.6

15

35.7

42

Side ? ‡

9

64.3

6.4

5

35.7

12

99

78

10.1

28

22

127

39

84.8

9.9

7

15.2

46

‡

Head injury without LOC †

Total head injuries Side +* †

Side -

38

71.7

10.5

15

28.3

33

Side ? ‡

22

78.6

9.9

6

21.4

28

*Consistency between head injury and pain side. † No consistency between head injury and pain side. ‡ Case for whom no reliable data were available regarding the site of injury, the injury was bilateral, or the pain of CH was not always located on the same side. CH—cluster headache; LOC—loss of consciousness.

were those with chronic CH that evolved from episodic (45.0% vs 25.0%). Dependable data regarding the site and date of head trauma with respect to the date of CH onset were available for 127 of the 138 patients who had a previous head injury; in 99 of 127 cases (78.0%), the head trauma occurred before CH, preceding it by an average of 10.1 years. Considering only the patients who had lost consciousness after sustaining a head injury, the trauma preceded CH in 40 of 44 patients (90.9%) and it did so by an average of 12.0 years. In the patients who were observed, there was no sure correspondence between the trauma site and the side where CH would develop later (Table 1). In 2000, Torelli et al. [31], after reviewing a personal case series of 69 patients with chronic CH, reported that those with chronic CH that evolved from episodic had a head injury in their medical histories more frequently (17/31 [54.8%]) than did those with chronic CH that was unremitting from onset (13/38 [34.2%]). Head trauma preceded CH onset in 64.7% of the patients with chronic CH that evolved from episodic (11/17) and in 76.9% of the patients with chronic CH that was unremitting from onset (10/13). Torelli et al. [32•], in an attempt to identify possible predictive factors for CH evolution from episodic to chronic, also conducted a comparative study of two groups of patients. The first group consisted of 28 CH patients (20 men and eight women) with chronic CH that evolved from episodic; the second group was comprised of 258

CH patients (185 men and 73 women) with a form of CH that remained episodic after the average number of years it had taken for CH to evolve from episodic to chronic in the first study group. With regard to the patients’ lifestyle habits and medical histories, differences emerged between the two groups in smoking and head trauma. For example, the latter was significantly more frequent in the group with chronic CH that evolved from episodic (Table 2). In most cases, head injury preceded the onset of CH and its evolution to the chronic stage. The average latency period between head injury and onset of CH was comparable in the two patient groups in terms of injuries without loss of consciousness (10.4 and 10.7 years, respectively). In head-injured patients who had suffered loss of consciousness, the average latency period was shorter in patients with chronic CH that evolved from episodic (5.5 years) compared with the other patients (10.7 years). Men with chronic CH that evolved from episodic were more likely than men with episodic CH to suffer a second head injury (4/28 and 5/258, respectively); in this case, the injury always preceded the evolution of CH to the chronic stage and it did so by many years. In 2001, Giglio et al. [33] reported a history of head trauma (with no known data regarding its severity and location) in 30 (27.8%) of the 108 CH patients in their personal case series (93 men and 15 women), including 85 patients with episodic CH, 14 with chronic CH, and nine with CH periodicity undetermined. Overall, the medical literature on the association between head trauma and CH includes a combination

134 Cluster Headache

Table 2. Head injury, loss of consciousness, and consistency between site of injury and pain side in chronic cluster headache evolved from episodic and episodic cluster headache patient groups* CCHE Female

ECH

Male

Total

Female

Male

Total

N

%

N

%

N

%

N

%

N

%

N

%

With LOC

1

12.5

5

25

6

21.4

6

8.2

30

16.2

36

14

Without LOC

3

37.5

8

40

11

39.3

11

15.1

39

21.1

50

19.4

Total

4

50

13

65

17

60.7

17

23.3

69

37.3

86

33.3

No injury

4

50

7

35

11

39.3

56

76.7

116

62.7

172

66.7

Yes

3

37.5

5

25

8

28.6

3

4.1

22

11.9

25

9.7

No

1

12.5

6

30

7

25

10

13.7

23

12.4

33

12.8

2

10

2

7.1

4

5.5

24

13

28

10.8

Headache injury

Injury site (same as pain side)

NA† No injury Total

4

50

7

35

11

39.3

56

76.7

116

62.7

172

66.7

8

100

20

100

28

100

73

100

185

100

258

100

*CCHE vs ECH (total sample) yes vs no: chi-square = 7.07, P < 0.008; CCHE vs ECH (male sample) yes vs no: chi-square = 4.68, P < 0.03. † Patients who reported a headache switching sides over cluster periods. CCHE—chronic cluster headache evolved from episodic; ECH—episodic cluster headache; LOC—loss of consciousness.

of reports on single cases or very small-size series of patients who developed CH or a cluster-like headache very shortly after sustaining a head injury; there also are descriptions sometimes of very large CH case series in which the presence of remote head trauma in the patients’ medical histories was investigated alongside other clinical and epidemiologic aspects. Five of the seven cases reported in the literature about CH occurring after enucleation of the ipsilateral eye also are discussed in this paper, because surgery was needed as a result of head trauma. On the other hand, we can reasonably assume that the role played by head trauma may be predominant over that of eye enucleation. If one compares the causes that led to eye enucleation in 112 patients, none of whom developed CH, who were observed by Sörös et al. [34] versus the seven patients with CH reported by other authors [20,22–24,30], it can be seen that head trauma was involved in only 35.7% of the former cases and in as much as 71.4% of the latter. Therefore, we think that two different aspects should be considered and discussed separately. The first is the possibility that head trauma may be a direct cause of CH, and the second is the presence or absence of a significant rate of head injuries in CH patients’ medical histories and the possible nature of this relationship.

Post-traumatic Cluster Headache The category of post-traumatic CH includes all of those headache forms that meet the diagnostic criteria of CH

and occur in close temporal relation to a head trauma. By “close temporal relation,” we should mean that CH must occur within 7 days of head trauma, according to the criteria set by the International Classification of Headache Disorders Second Edition (ICHD-II) [2] for post-traumatic headache. In the introductory section to group 5 “Headache attributed to head and/or neck trauma,” ICHD-II states that “When a new headache occurs for the first time in close temporal relation to a known trauma, it is coded as secondary headache attributed to the trauma. This is also true if the headache has the characteristics of migraine, tension-type headache, or cluster headache.” Post-traumatic headache sometimes can be very similar but not entirely identical to CH. That is the reason why ICHD-II adds that “A variety of pain patterns may develop after head injury, and may closely resemble primary headache disorders… In some cases…a cluster-like syndrome has been described in a few patients.” The ICHD-II also clearly states in the diagnostic criteria of post-traumatic headache that “Headache develops within 7 days after head trauma.” If we keep these ICHD-II criteria in mind, then the only sure case of post-traumatic CH reported in the literature is that described by Turkewitz et al. [28•], who presented a patient in whom CH occurred 6 days after head trauma. Strictly speaking, neither the four patients who developed CH after a few weeks described by Reik [25], nor the patient in whom CH appeared only after 45 days described by Formisano et al. [27] can be ascribed to this category. Among the cases reported in the literature with CH onset after eye enucleation, the one with the

Head Trauma and Cluster Headache Manzoni et al. 135

shortest latency between the two events is the patient who began to suffer from CH 3 weeks after surgery (and nearly 8 months after head trauma) described by Evers et al. [30]. One of the cases with eye enucleation reported by Rogado and Graham [20] began to suffer from CH shortly after a mild head injury, which had been preceded by another, more severe injury that had led to eye enucleation 4 years before. Therefore, in this case, the causal relationship between the last milder trauma and CH is very doubtful. Unfortunately, other literature reports do not provide any information about latency between head trauma and CH onset [5,16,18,26]. Post-traumatic CH seems to be a very rare occurrence, which is similar to other forms of secondary CH [15].

Head Trauma/Cluster Headache Association Several case series [5,13••,17–19,21,29,33], described over a very long period of time, report a significant association between head trauma and CH. Should there be any doubt that a recall bias in CH patients may negatively affect the reliability of data, the fact remains that some of the studies [13••,21,29] compared CH patients with a control group made up of patients with other forms of primary headache. In any event, the presence of head injuries in the medical histories of CH patients is so frequent (approximately one third of patients in the largest, best-studied case series) that there can be few doubts regarding the not merely casual nature of the association between the two events. Although more specific and more methodologically correct studies are needed to clarify the matter, based on current knowledge we can state that, in all likelihood, there exists a significant association between head trauma and CH. By contrast, the nature of this association remains uncertain. Some literature findings may help us advance a few hypotheses. The first finding regards the latency between head trauma and CH onset, which was 5 and 25 years, respectively, in the two cases reported by Symonds [17], 2 to 30 years in the four cases reported by Lance and Anthony [19], and an average of 9 to 14 years in the case series reported by Manzoni et al. [13••,21]. Therefore, the interval separating the two events covers a broad time range. The second finding is that there is not always a correspondence between the trauma site and the CH side [21]. The third finding is that the extent of head trauma does not appear to play a fundamental role. Based on literature evidence, the first possible interpretation of the head trauma/CH relationship is that the trauma may cause damage to extra- or intracranial peripheral or central nervous structures, thus predisposing to the future development of CH. According to Lance and Anthony [19] and to Sjaastad [35], the fact that CH occurs more frequently in men than in women may depend on the greater vulnerability of men to traumatic events.

The second possible interpretation is that the lifestyle or personality of subjects who are bound to develop CH over time may in itself be a risk factor for head trauma [36]. It has been known for some time and reported by several authors [5,21,37] that CH patients tend to overindulge in some non-essential habits, such as smoking, alcohol, or coffee intake. They also may be inclined to adopt risky behaviors that make them more prone to accidents (eg, while driving or working). A detailed analysis, which is currently near completion, of the ways in which head trauma occurred in the CH patients followed at our headache center seems to indicate that head trauma was significantly more frequent in CH patients than in control subjects after motor vehicle accidents in which they were the responsible party. Although not excluding the first interpretation proposed about the relationship between head trauma and CH, this last finding seems to lend more support to the second interpretation. Another interesting hypothesis is that advanced by Torelli et al. [32•,38] that head injuries may represent an unfavorable prognostic indicator for CH, favoring the evolution of episodic CH into chronic CH.

Conclusions Among the major forms of primary headache, CH is the one with the most distinctive clinical features, which are repeatable between attacks in the same patient and from one patient to the next. Thanks to the description of large case series of CH patients by several authors [5,21], the clinical picture of CH is now very well defined. However, some clinical and epidemiologic aspects reported during the past few decades have not been confirmed in more goal-specific and methodologically appropriate studies or have not been easily interpreted and still remain unexplained. For example, the association of a duodenal ulcer and CH, reported in the 1970s and 1980s [5], was not found to be statistically significant during recent investigations [29]. The increased mortality for cancer, especially lung cancer, reported by Kudrow [5] in CH patients appears to be a consequence of the heavy smoking habit of most of these patients. Other aspects, such as the increased frequency of CH in men, the high number of heavy smokers among affected patients, and the statistically significant association of CH with head trauma, remain unclear. The nature of the relationship between head trauma and CH (ie, head injuries induce damage potentially leading to the future development of CH, or head injuries fall within those lifestyle-related risks that are typical of most CH patients) will be explained only through specific epidemiologic research. The results of these studies certainly will help to clarify the etiopathogenetic mechanisms of this particular type of primary headache.

136 Cluster Headache

References and Recommended Reading Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance

Kunkle EC, Pfeiffer JB Jr, Wilhoit WM, Hamrich LW: Recurrent brief headaches in “cluster pattern.” Trans Am Neurol Assoc 1952, 56:240–243. 2. Headache Classification Subcommittee of the International Headache Society: The International Classification of Headache Disorders, 2nd Edition. Cephalalgia 2004, 24(suppl 1):1–151. 3.• Torelli P, Manzoni GC: Pain and behaviour in cluster headache: a prospective study and review of the literature. Funct Neurol 2003, 18:205–210. An interesting study about the distinctive aspects characterizing CH patients. 4. D’Alessandro R, Gamberini G, Benassi G, et al.: Cluster headache in the Republic of San Marino. Cephalalgia 1986, 6:159–162. 5. Kudrow L: Cluster Headache: Mechanisms and Management. New York: Oxford University Press; 1980. 6. Sjaastad O, Bakketeig LS: Cluster headache prevalence: Vågå study of headache epidemiology. Cephalalgia 2003, 23:528–533. 7. Torelli P, Beghi E, Manzoni GC: Cluster headache prevalence in the Italian general population. Neurology 2005, 64:469–474. 8. Manzoni GC: Gender ratio of cluster headache over the years: a possible role of changes in lifestyle. Cephalalgia 1998, 18:138–142. 9. Goadsby PJ: Pathophysiology of cluster headache: a trigeminal autonomic cephalgias. Lancet Neurol 2002, 4:251–257. 10. May A: Cluster headache: pathogenesis, diagnosis, and management. Lancet 2005, 366:843–855. 11. Russell MB, Andersson PG, Thomsen LL: Familial occurrence of cluster headache. J Neurol Neurosurg Psychiatry 1995, 58:341–343. 12. Rainero I, Rivoiro C, Gallone S, et al.: Lack of association between the 3092 T�C Clock gene polymorphism and cluster headache. Cephalalgia 2005, 25:1078–1081. 13.•• Manzoni GC: Cluster headache and lifestyle: remarks on a population of 374 male patients. Cephalalgia 1999, 19:88–94. An article investigating in detail the presence of previous head injuries and non-essential habits such as smoking and alcohol intake in a large case series of male patients with CH. 14. Levi R, Edman GV, Ekbom K, Waldenlind E: Episodic cluster headache II: high tobacco and alcohol consumption in males. Headache 1992, 32:184–187. 15. Carter DM: Cluster headache mimics. Curr Pain Headache Rep 2004, 8:133–139. 16. Hunter CR, Mayfield FH: Role of the upper cervical roots in the production of pain in the head. Am J Surg 1949, 78:743–749. 17. Symonds C: A particular variety of headache. Brain 1956, 79:217–232. 18. Friedman AP, Mikropoulos HE: Cluster headaches. Neurology 1958, 8:653–663. 1.

Lance JW, Anthony M: Migrainous neuralgia or cluster headache? J Neurol Sci 1971, 13:401–414. 20. Rogado AZ, Graham JR: Through a glass darkly. Headache 1979, 19:58–62. 21. Manzoni GC, Terzano MG, Bono G, et al.: Cluster headache: clinical findings in 180 patients. Cephalalgia 1983, 3:21–30. 22. McKinney AS: Cluster headache developing following ipsilateral orbital exenteration. Headache 1983, 23:305–306. 23. Bond JB, Wesley RE: Cluster headache after orbital exenteration. Ann Ophthalmol 1987, 19:438. 24. Prusinski A, Liberki PP, Szulc-Kuberska J: Cluster headache in a patient without an ipsilateral eye. Headache 1985, 25:134–135. 25. Reik L Jr.: Cluster headache after head injury. Headache 1987, 27:509–510. 26. Mathew NT, Rueveni U: Cluster-like headache following head trauma. Headache 1988, 28:297. 27. Formisano R, Angelini A, De Vuono G, et al.: Cluster-like headache and head injury: case report. Ital J Neurol Sci 1990, 11:303–305. 28.• Turkewitz LJ, Wirth O, Dawson GA, Casaly JS: Cluster headache following head injury: a case report and review of the literature. Headache 1992, 32:504–506. The article reports the only sure case of post-traumatic CH in the literature. 29. Italian Cooperative Study on the Epidemiology of Cluster Headache: Case-control study on the epidemiology of cluster headache I: etiological factors and associated conditions. Neuroepidemiology 1995, 14:123–127. 30. Evers S, Sörös P, Brilla R, et al.: Cluster headache after orbital exenteration. Cephalalgia 1997, 17:680–682. 31. Torelli P, Cologno D, Cademartiri C, Manzoni GC: Primary and secondary chronic cluster headache: two separate entities? Cephalalgia 2000, 20:826–829. 32.• Torelli P, Cologno D, Cademartiri C, Manzoni GC: Possible predictive factors in the evolution of episodic to chronic cluster headache. Headache 2000, 40:798–808. This article reports a number of new, interesting indications about possible negative prognostic indicators for CH. In particular, the possibility is discussed that head trauma may favor the evolution of episodic CH to chronic CH. 33. Giglio JA, Bruera OC, Figuerola ML, Leston JA: Cluster headache: clinical features and habits in La Plata. Cephalalgia 2001, 21:494. 34. Sörös P, Vo O, Gerding H, et al.: Enucleation and development of cluster headache: a retrospective study. BMC Neurol 2005, 5:6–10. 35. Sjaastad O: Cluster Headache Syndrome. London: WB Saunders; 1992. 36. Ekbom K: Lifestyle factors in males with cluster headache [Editorial commentary]. Cephalalgia 1999, 19:73–74. 37. Graham JR: Cluster headache. In Pathogenesis and Treatment of Headache. Edited by Appenzeller O. New York: Spectrum Publications, Inc.; 1976:93–108. 38. Torelli P, Manzoni GC: What predicts evolution from episodic to chronic cluster headache? Curr Pain Headache Rep 2002, 6:65–70. 19.