Etiopathogenesis of hyperostosis frontalis interna: A mystery still

Page 1

E

AAa

b

c

a

ARRA

KCDFHOP

1

phifilaSMiosai

M4

(ms

0d

Annals of Anatomy 193 (2011) 453– 458

Contents lists available at ScienceDirect

Annals of Anatomy

jo ur n al ho mepage: www.elsev ier .de /aanat

tiopathogenesis of hyperostosis frontalis interna: A mystery still

thanasios Raikosa,b,∗ , George K. Paraskevasb, Faisal Yusufa , Panagiota Kordalib , Soultana Meditskouc,bdulatif Al-Haja, Beate Brand-Saberia

Department of Anatomy and Molecular Embryology, Medical Faculty, Ruhr University, Bochum, GermanyDepartment of Anatomy, Medical School, Aristotle University of Thessaloniki, GreeceLaboratory of Histology, Embryology and Anthropology, Medical School, Aristotle University of Thessaloniki, Greece

r t i c l e i n f o

rticle history:eceived 4 January 2011eceived in revised form 28 April 2011ccepted 7 May 2011

eywords:

a b s t r a c t

Hyperostosis frontalis interna is a morphological pattern characterized by single or multiple bony nod-ules situated on the inner lamina of the frontal bone. It is seldom found in males, but it is a commonphenomenon among post-menopausal females in modern societies but relatively rare in antiquity. Theetiopathogenesis of the trait is a matter of debate and ranges from genetic predisposition to epigenetic,while endocrine disturbances, aging, and dietary factors are also listed among the causes. We studied thefrequency, characteristic features, and etiopathogenesis of the disease in recent cadaveric and dry skull

adaverura materalx cerebriyperostotic nodulesssificationathogenesis

specimens. The frequency of hyperostosis frontalis interna in cadavers and dry skull materials was almostidentical, 12.5% and 12.3%, respectively. In cadavers, 87.5% of severe hyperostosis frontalis interna caseswere found in females over 65 years-old. Interestingly, in two cadavers we found hyperostotic lesionsspreading onto adjacent tissues such as the dura and falx cerebri. We provide some new aspects thatmay help in better understanding of the etiopathogenesis of hyperostosis frontalis interna. Thereby, wediscuss the various etiopathogenesis models found in the literature.

. Introduction

Hyperostosis of the cranium refers to a set of endocranialathological features, each having an independent etiology, variedistopathological findings, and localization. Hyperostosis frontalis

nterna (HFI) is characterized by the presence of a single or, morerequently, multiple nodules localized bilaterally on the inner lam-na of the frontal bone, while most of the diploë and externalamina of the bone remain unaffected. The calvarial midline is usu-lly spared even in the most severe cases (Moore, 1936a; She andzakacs, 2004; Talarico et al., 2008). HFI was firstly described byorgagni in 1719 as a pathology accompanying obesity and viril-

sm (Morgagni, 1719). HFI can be encountered in relation to manyther syndromes such as: Troell–Junet (acromegaly, thyrotoxico-

is), Stewart–Morel (mental disease, obesity), Paget’s (enlargednd deformed bones), Frölich (obesity, growth retardation, pitu-tary hypocrinism), and Klippel–Trenaunay–Weber (varicose veins,

Abbreviation: HFI, hyperostosis frontalis interna.∗ Corresponding author at: Department of Anatomy and Molecular Embryology,edical Faculty, Ruhr University of Bochum, Geb. MA 5/158, Universitätsstr. 150,

4801 Bochum, Germany. Tel.: +49 0234 32 27780.E-mail addresses: [email protected] (A. Raikos), g [email protected]

G.K. Paraskevas), [email protected] (F. Yusuf), [email protected] (P. Kordali),[email protected] (S. Meditskou), [email protected] (A. Al-Haj), beate.brand-

[email protected] (B. Brand-Saberi).

940-9602/$ – see front matter © 2011 Elsevier GmbH. All rights reserved.oi:10.1016/j.aanat.2011.05.004

© 2011 Elsevier GmbH. All rights reserved.

port-wine stain, bone and soft tissue hypertrophy) (Anton, 1997;Hershkovitz et al., 1999). However, it was acknowledged that itcan also occur as an independent entity (Caughey, 1958; She andSzakacs, 2004).

The normal calvarium is composed of cortical bone in theexternal lamina (exocranium), compact bone in the inner (endocra-nium), while trabecular bone lies between them (diploë). Thereis no statistical difference between males and females in diploicthickness, except the frontal bone in males. Furthermore, no cor-relation exists between diploic thickness and age, height or weight(Lynnerup et al., 2005). HFI is usually encountered more frequentlyin elderly females (Henschen, 1949; Moore, 1936a; Hershkovitzet al., 1999) and occurs rarely in males (Nikolic et al., 2010). Itseems that every reported case in males was related to severehypogonadism and testicular atrophy (Caughey, 1958; Perou, 1964;Yamakawa et al., 2006).

The severity of HFI varies greatly in shape and size, and can beclassified using radiologic (indirect) or histological (direct) meth-ods. Barber et al. (1997), designated radiological criteria and gradedthe appearance of HFI in plain radiographs: Grade I, featured byearly bone formation in the endocranium; Grade II, prominent bonethickness and roundish appearance of nodules on the inner calvar-

ial plate; while in Grade III, the thickness of the endocranium isextensive and shows an irregular pattern of nodules. Hershkovitzet al. (1999), in order to establish an accurate identification of theseverity of HFI, proposed a classification grouping the four types

Page 2

4 Anatomy 193 (2011) 453– 458

aeuf2tgocsm

2

fafiometdaiwwJ

piaisdeo

3

iftiaretchf1s(twshtp(it

ogy, while an increase in cancellous bone in the diploë was present.The marrow cavities along with bony trabeculae had an irregularshape and size pattern. The endocranial plate exhibited extensiveremodeling similar to that of the diploë, whereas the lamella of

54 A. Raikos et al. / Annals of

ccording to morphologic and histopathologic characteristics of thendocranium: Type A, single or multiple isolated bony elevationsnder 10 mm in diameter located on the endocranial surface of therontal bone; Type B, nodular bony formations occupying less than5% of the frontal bone; Type C, same as B but nodules covering upo 50% of the frontal bone; Type D, continuous nodular bony over-rowth involving over 50% of the frontal endocranium. The aimf our investigation has been to interpret HFI findings in dry andadaveric skulls of body donators, to comment on the pathogene-is of the various traits, and to advance our understanding of theodels proposed over the years.

. Materials and methods

For the purpose of this study, 204 dry skulls (107 male and 97emale) from our recent osteological collection (aged 20–94 yearst death), and 40 formalin embalmed cadavers (18 male and 22emale), were dissected with an oscillating bone saw, and exam-ned for the presence of irregular bone patterns on the inner laminaf the calvarium. Both cadaveric and dry skull bones belonged toembers of the Caucasian ethnic group from the central and south-

rn Europe region, and had been born during the 20th century. Inhe cadaveric specimens, special care was taken during calvarialissection and removal to preserve the underlying structures suchs the meninges and cerebrum. In addition, the meninges werenspected for the presence of ossification lesions and resected along

ith the brain in order to expose the cranial fossae. Measurementsere obtained using a Vernier digital sliding-caliper (Mitutoyo Co.,

apan), accurate to 0.1 mm.The severity of HFI was classified according to the four types pro-

osed by Hershkovitz et al. (1999). The specimens were classifiednto three age groups (18–39, 40–64, and >65), the young, middle,nd the old age group. Histological samples were taken bilaterallyn cases of macroscopic recognition of HFI with the use of narrowaw blades, then prepared and stained according to a previouslyescribed technique according to Talarico et al. (2008). The cadav-rs were free of any other bone related pathology, and no historyf brain tumor or meningioma was present.

. Results

The total frequency of HFI in both cadaveric and dry skull spec-mens was 11.9%. The frequency in cadavers was 12.5% (22.7% inemales, 0% in males). The presence in the older age group amonghe cadaveric specimens was 28.6% (Table 1), whereas the sever-ty of HFI in female cadavers and dry skulls is correlated withn increase in age (r = 0.99, and 0.97 for cadavers and dry skulls,espectively). 87.5% of Type D cases were over 65 years in age. Inter-stingly, in an 83-year-old female cadaver an irregular pattern ofhe inner frontal calvarial lamina was noted. The frontal bone wasovered bilaterally by continuous multiple small and medium sizedard nodules, organized in groups. During the inspection of the

alx cerebri, a hard dense sphenoid-shaped lesion 1.9 cm long and.5 cm wide was noted, corresponding to the level of the coronaluture, whereas it transversed the dura from one side to the otherFig. 1). The borders were smooth and regular. Interestingly, his-ological study revealed that the composition of the dense massas bone with a pattern similar to that of HFI (Fig. 2). The pos-

ibility of epidural fibrous tumor or fibromatosis due to abscess orematoma was excluded due to the different histopathological pat-ern, localization and medical history. The irregular calvarial bone

attern could be sorted as Type D according to Hershkovitz et al.1999) classification model, however, the rare hyperostotic find-ng at the falx cerebri could not be classified to any of the knownypes and was designated as Type E (Table 1). In a second case, in

Fig. 1. Hard dense sphenoid-shaped lesion (arrows) 1.9 cm long and 1.5 cm wide onthe falx cerebri at the level of the coronal suture. The nodule perforates the dura onboth sides.

a 79-year-old female with severe HFI (Type D), the dura coveringthe frontal bone was ossified and elevated on the left of the ante-rior cranial fossa (Fig. 3a and b). The inferior segment of the frontallobe was squeezed as a result of the evoked compression (Fig. 4).The frequency of HFI in female dry skulls was 20.6%, while onlythree cases were found among the 107 male dry skulls. In males,two of the cases belonged in the old and one in the middle agedgroup (Table 2). The severity of the trait in males was limited onlyto Types A and B.

The hyperostotic nodules were round or oval shaped and theircaliber ranged from a few millimeters to 14 mm, while the pro-trusions from the physiological border of the endocranium rangedfrom 1 mm to 19 mm. The midline of the horizontal portion of thecalvarium was spared in all the studied cases, however in severecases the vertical portion was occupied by hyperostotic lesions(Fig. 5). Some small roundish and/or irregularly shaped noduleswere noticed on the inner lamina of calvarial bones other than thefrontal (usually the parietal bone) in 3 out of 29 of total specimenswith HFI (10.3%) (Fig. 6). In all cadaveric specimens, the dura materhad the tendency to adhere in hyperostotic areas, while in someinstances it was firmly attached between nodule islets (Fig. 5).

Histopathological studies of the HFI nodules indicated that theexternal lamina of the frontal bone presented a typical morphol-

Fig. 2. Histological section of the bony nodule perforating the falx cerebri. Bonemarrow cavities and bony trabeculae show extensive remodeling. Ectocranium withdense bone. Increase in cancellous bone in the diploë, the marrow cavities have anirregular shape and size pattern (arrowheads). Magnification: 40×, paraffin sectionat 6 �m, stain: H + E.

Page 3

A. Raikos et al. / Annals of Anatomy 193 (2011) 453– 458 455

Table 1Frequency and classification of hyperostosis frontalis interna in cadavers. Types A–D is based on Hershkovitz et al. (1999) classification. Type E corresponds to severehyperostosis frontalis interna with soft tissue expansion. n: number of cadavers.

Cadavers (n:40) TotalType A Type B Type C Type D Type E

Females (n:22)18–39 (n:2, 9%) – – – – – 040–64 (n:6, 27.3%) – – 1 – – 1 (16.7% in this age group)

5 (22.7% in females)>65 (n:14, 63.7%) 1 – – 1 2 4 (28.6% in this age group)Males (n:18)18–39 (n:1) – – – – – 0 040–64 (n:1) – – – – – 0>65 (n:16) – – – – – 0

Total 1 0 2 2 1 5 (12.5%)

Table 2Frequency of hyperostosis frontalis interna in modern era dry skulls. Types A–D is based on Hershkovitz et al. (1999) classification. n: number of cadavers.

Dry-skulls (n:204) Total

Type A Type B Type C Type D

Females (n:97)18–39 (n:12, 12.4%) 1 – – – 1 (8.3% in this age group)

22 (22.7% in females)40–64 (n:21, 21.6%) 1 1 – 1 3 (14.3% in this age group)>65 (n:64, 66%) 3 5 6 4 18 (28.1% in this age group)Males (n:107)18–39 (n:18, 16.8%) – – – – 0

3 (2.8% in males)40–64 (n:30, 28%) 1 – – – 1 (3.3% in this age group)>65 (n:59, 55.2%) 1 1 – – 2 (3.9% in this age group)

Total 7 7 6 5 25 (12.3%)

Fig. 3. (a and b) Blade-like ossified dura (arrows) mostly at the left side of the frontalbone (a) and at the anterior cranial fossa (b). The lesion was tethered to the duramater covering the anterior cranial fossa as well as the frontal lobe. The ossified duracoursed horizontally to the anterior border in contact with the inferior surface of thefrontal lobe of the brain while the posterior surface was free. The lesion terminatedon the vertical segment of the frontal bone.

Fig. 4. The inferior surface of the left frontal lobe was affected by the presence ofossified and elevated dura, hence the medial orbito-frontal (arrows), straight (St),superior frontal (S), middle (M), and inferior (I) frontal gyri are found to have mildatrophy and depressions due to compression phenomena.

Page 4

456 A. Raikos et al. / Annals of Anatomy 193 (2011) 453– 458

Fig. 5. Severe form (Type D) of hyperostosis frontalis interna. The midline of thehorizontal portion (Mh) of the calvarium is free of any pathology, but the midlineoTi

tctn

4

sYcrt(oaH

utpFobsit

Fn

Fig. 7. Panoramic view of a typical histological specimen of hyperostosis frontalisinterna. The cancellous bone in diploë is present. Ectocranium (arrows) with densebone. The marrow cavities (arrowheads) and bony trabeculae have an irregularshape and size pattern. The endocranial plate appears to have undergone extensiveremodeling similar to that in diploë, whereas compact bone was substituted by irreg-ularly shaped marrow cavities and trabeculae. Gaps can be seen in the endocranial

f the vertical portion (Mv) of the frontal bone is occupied by hyperostotic lesions.he dura had the tendency to adhere in hyperostotic areas, while in some instancest was firmly attached between nodule islets (arrowheads).

he compact bone was substituted by irregularly shaped marrowavities and trabeculae (Fig. 7). Moreover, gaps were noticed inhe endocranial surface. In all the studied samples, no evidence ofeoplastic cells was found.

. Discussion

Initially, HFI was found to be associated with obesity, hir-uitism and virilitism in Morgagni syndrome (Morgagni, 1719).ears later, other authors have associated it to various neuropsy-hiatric disorders such as headache and discussed its possibleelation to endocrine imbalance, while the present terminology forhe above complex is Morgani–Stewart–Morel–Moore syndromeHajdu et al., 2009). However, HFI can be found unaccompanied byther features of any given syndrome and thus can be consideredn independent phenomenon (Moore, 1935, 1936a,b; Perou, 1964;ershkovitz et al., 1999; Talarico et al., 2008).

HFI usually does not induce any symptoms, but, if the bony nod-les protrude extensively (bullet-like) or they become too large,he underlying soft tissues such as dura and cerebrum can be com-ressed or impinged (Chaljub et al., 1999; Talarico et al., 2008).urthermore, symptoms may be evoked related to compressionf the cortex area involved. It is well known that chronic cere-

ral compression can lead to atrophy (Devriendt et al., 2005), inuch a case the surgical excision of the prominent nodules in thenner lamina of the endocranium is the only means of treatmento relieve the symptoms (Hasegawa et al., 1983). Neuropsychiatric

ig. 6. Small roundish (arrows) and irregularly shaped nodules (arrowheads) wereoticed in parallel to the midline in the internal lamina of parietal bone.

surface (double arrows). Increase in cancellous bone in the diploë. Magnification:40×, paraffin section at 6 �m, stain: H + E.

symptoms, epilepsy, dementia, and headache can be encounteredin patients with nodular endocranial remodeling. In a study onelderly patients with various forms of hyperostosis cranii, includingHFI, the authors were able to identify transient generalized spikeand slow wave complexes over the frontal lobes of a patient’s elec-troencephalography evoked by nodular lesions in the frontal andparietal bone (Sohmiya et al., 2001). Behavioral disturbances andneed for psychiatric care was the prominent feature in all accessedmedical histories of 13 HFI cases on an extensive autopsy study(Devriendt et al., 2005). On the contrary, in an older study, vari-ous neurophysiologic, metabolic, radiologic, and endocrine studieswere performed in a series of 11 female psychiatric patients withHFI. The results failed to suggest a direct relationship of HFI tobehavioral disorders, suggesting that the co-existence of nodularlesions in X-rays was a coincidental finding (Smith and Hemphill,1956).

HFI is usually an incidental finding in X-ray, CT and MRI stud-ies, while milder cases are more difficult to distinguish in plainradiography (Barber et al., 1997). In one study conducted on 65patients with HFI, the pathology has been associated with highserum alkaline phosphatase in 53% of them. However, no clear clin-ical significance has been provided for the findings (Gegick et al.,1973).

4.1. Frequency of HFI in historic and modern era population

The frequency of HFI differs greatly between studies conductedin the modern era or on the population prior to the 19th century.In an extensive study conducted on 2019 pre-19th century skulls,Hershkovitz et al. (1999) did not find even a single case of HFI,whereas in an early 20th century collection of 1706 skulls, 219

cases (12.8%) of HFI were identified, with the relative frequency inwomen and men being 24% and 5.2%, respectively. The severity ofHFI is age related, while the most frequently recorded Types C andD cases are found in postmenopausal women (Hershkovitz et al.,

Page 5

Anato

1bff2a

ow(lhatlef

icqpmaouptfdat

aga(etH2r1(

4

alHiba(HwropsceIf

A. Raikos et al. / Annals of

999; Nikolic et al., 2010). The frequency in a study on 803 skullsones (412 male and 391 female) in a series of skeletal collectionsrom 14th century BC to 17th AD was 1.21% and 3.83% in males andemales, respectively, thus a total frequency of 2.49% (Hajdu et al.,009). The frequency in males is higher in clinical samples than inntique skeletons (Rühli et al., 2004).

However, in one study performed on an osteological collectionf Ancestral Pueblans (Anasazi) the frequency of HFI in femalesas 44% and 8.3% in males, while the overall frequency was 32.4%

Mulhern et al., 2006). Surprisingly, the frequency rate is very simi-ar to those studies conducted on the last two centuries population,ence we should underline that the ancient Pueblo Bonito was

residence, mostly for very wealthy people. In order to explainhe phenomenon we can hypothesize that the eating habits andifestyle of these people had many similarities to those of the mod-rn populations, supporting the hormonal model as the genesisactor of HFI.

Additionally, some interesting conclusions have been publishedn an effort to explain the higher frequency of HFI in modernompared to historical/ancient population as well as the higher fre-uency in females. A study claims that in many excavation sites,arts of the calvarium and the pelvic bones were missing post-ortem, thus leading to an unrealistic overall number of cases

long with inaccurate determination of the female skeletons. More-ver, the shorter lifespan of the ancient populations cannot benderestimated (Armelagos and Chrisman, 1988). Plain radiogra-hies are ineffective for detection of early HFI stages as opposedo naked eye inspection. However, even small single lesions of aew millimeters in gauge can be distinguished in cadaveric andry skulls (Hajdu et al., 2009). Additionally, the great droughts inncient periods potentially affected the living conditions and nutri-ion of that time.

In males, the presence of the trait is 4–5 times less frequentnd it is usually related to testicular atrophy and decreased andro-ens in advanced age (Hershkovitz et al., 1999). However, therere rare reports of HFI such the one found in a 35-years-old manClaassen, 1989). When present in men, HFI is usually moderate inxtent (usually Types A and B according to Hershkovitz classifica-ion), while Type D is rarely reported. Cadaveric studies regardingFI are limited in the literature. In an autopsy study, conducted on48 females and 1220 males, the frequency was 18.1% and 0.74%,espectively (Nikolic et al., 2010). In a second study, carried out on532 cadavers, the HFI was found in females was 0.78% and 0.065%only a single case) in males (Devriendt et al., 2005).

.2. Explanatory models of HFI

Since the discovery of the trait, the exact etiology of HFI remains mystery, however thenceforth many researchers have postu-ated that hormonal disturbances are the predominant factor forFI genesis. It seems that repeated dysendocrinism through life,

nvolving and not limited only to, estrogens but on physiologicalalance of estrogens, progesterone, androgens, and maybe leptinnd dietary phytoestrogens is possibly the initiative cause of HFICalame, 1951; Hershkovitz et al., 1999; Morel, 1930; Rühli andenneberg, 2002). Besides, there are studies correlating obesityith increased presence of HFI (Gegick et al., 1973). In a series

eport on 17 patients with HFI, all individuals were females, 82%f the patients were obese with an average weight excess of 50ounds each, while 12 of them suffered from endocrine diseasesuch as amenorrhea, early or artificial menopause, sterility, and

ystic ovaries. Four patients were diagnosed with thyroid relatedndocrinopathies and one with diabetes mellitus (Harding, 1949).nterestingly, lesions similar to those encountered in HFI wereound after estrogen administration in mice (Rudali, 1968).

my 193 (2011) 453– 458 457

Over the years, several proposals have been discussed as to thepossible pattern of bone overgrowth in HFI. Three major modelsdominate so far: the American, the European, and the Global. TheAmerican model, proposed by Moore in 1955 suggested that thereis an increased diploic volume due to spongy bone proliferationwhile the outer lamina of the frontal bone remains unaffected as aresult of its physical properties (Moore, 1955). The European modelproposes that the genesis of the nodular lesion is caused by aggran-dized intradural vasculature and adheres to the concept that durais the initiating site of HFI (Thevoz, 1966). The global model pro-poses a four stage process (Hershkovitz et al., 1999). Osteogeniccells form new layers of lamellar bones which are deposited on thedura followed by neovascularisation and finally sclerotic solidifica-tion leaving irregular cavities and gaps. The large cavities providethe effect of the indented inner plate of the endocranium. Lastly,the newly formed bone expands towards the diploic space.

In one of our studied specimens, we noticed a bony lesion sit-uated at the falx cerebri (Fig. 1). While in a second one, the durapresented as a blade-like ossified appearance on the frontal bone(Fig. 3a) and in the anterior cranial fossa (Fig. 3b). We believe thatthe global model provides an adequate explanation for both caseswith the very rare spread to the adjacent soft tissues reinforcing thetheory. To the best of our knowledge, there have been no previousreports in the literature associating HFI with hyperostotic lesionslocated at the falx cerebri and dura mater ossification.

4.3. Etiopathogenesis of HFI

The main controversy lies in the etiopathogenesis of HFI whilemany models have been suggested by many medical specialtiesas well as anthropologists. Although most studies do not endorse agenetic predisposition to the trait, there have been reports on skullswith familiar affinity (Glab et al., 2006; Hajdu et al., 2009). A sep-arate angiosome composed of numerous vascular anastomoses onthe frontal aspect of the dome of the skull satisfactorily explains thelocalization of HFI in the frontal bone of the calvaria (Perou, 1964).Moreover, supporting the estrogen-theory, the primary ossifica-tion centers located on each frontal bone are re-initiated by thepresence of estrogens. However, it is well known that the estrogenlevel remains relatively constant for 25 years after menarche andthen gradually decreases until menopause. Some studies claim thatthe increased frequency of HFI in the modern era is connected tomicroevolutionary hormonal changes due to easier access to foodand higher metabolic rates and, thus, profoundly increased lep-tin levels (Rühli and Henneberg, 2002), although its effect on bonemetabolism is debated (She and Szakacs, 2004). We believe thatthe appearance of HFI in some rich and well nourished antiquitysocieties have mislead various researchers to hypothesize that HFIis a genetically transmitted phenomenon among specific historicalpopulations. In a study conducted on 259 females aged 60- to 80-years-old, 127 HFI cases were detected, while of these almost 3/4of them were obese (Verdy et al., 1978).

The consumption of dietary phytoestrogens from sources suchas soy, grains, linseed and vegetables has increased over the lastdecades. Estrogen metabolism can be affected, and bone densitycan be benefited by these food sources pre- and post-menopausally(Cassidy and Faughnan, 2000). The estrogenic effect of these foodsources is dose-dependent, thus in societies with a high consump-tion of phytoestrogens the frequency of HFI may be greater thanthose with a more mixed food life style. This may help to explainthe differences in frequency among historic populations and clar-ify assumptions about HFI and genetic predisposition. However, the

complexity of interactions between environmental and biologicalfactors still needs to be investigated (Mulhern et al., 2006).

An interesting hypothesis that could add to the further explana-tion of HFI etiopathogenesis in males, collaterally also to females,

Page 6

4 Anato

chemctstfa

f1HttisaetcititnH

C

A

f

R

A

A

B

CC

C

C

C

D

58 A. Raikos et al. / Annals of

oncerns chronic androgenic steroid abusers. These individualsave very high androgen levels but due to aromatization of thexcessive hormonal balance to estrogen they often develop gyneco-astia as well as testicular atrophy (Hall and Hall, 2003). After

essation of the exogenous steroid administration, in many cases,he testicular atrophy is resistant or irreversible even after the pre-cription of gonadotropins (van Breda et al., 2003). In the future,he researchers pondering the topic should evaluate whether therequency of HFI is statistically higher in chronic androgen abusingthletes compared to healthy individuals of the same age group.

There are reports correlating hyperprolactinemia with a highrequency of HFI (Fulton et al., 1990; Pawlikowski and Komorowski,987). As mentioned before, dystrophia myotonica is related toFI, is distinguished by gonadal atrophy in both genders, while

he diminished levels of urinary 17-ketosteroids in males indicateshe malfunction of adrenal glands and testicles (Caughey, 1958). Its suggested that the anterior pituitary gland, in an effort to rein-tate proper androgenic function, my oversecrete gonadotropinsnd growth hormone. Hence the increased growth hormone lev-ls may evoke HFI. However, one may expect all calvarial boneso be affected by the growth hormone as is the case in metabolicraniopathy, but this hypothesis cannot be excluded until furthernvestigation has been done. HFI was a very rare phenomenon prioro the 19th century (Hershkovitz et al., 1999). It is assumed that thiss due to different patterns of menarche and menopause accordingo social behavior of older periods. The frequent pregnancies andursing kept estrogen levels low, thus acting negatively on possibleFI genesis.

onflict of interest

The authors declare that they have no competing interests.

cknowledgement

The authors would like to acknowledge Ms. Claudia Schneideror her excellent technical support in cadaveric dissection.

eferences

nton, S.C., 1997. Endocranial hyperostosis in Sangiran 2 Gibraltar 1, and Shanidar5. Am. J. Phys. Anthropol. 102 (1), 111–122.

rmelagos, G.J., Chrisman, O.D., 1988. Hyperostosis frontalis interna: a Nubian case.Am. J. Phys. Anthropol. 76 (1), 25–28.

arber, G., Watt, I., Rogers, J., 1997. A comparison of radiological and palaeopatho-logical diagnostic criteria for hyperostosis frontalis interna. Int. J. Osteoarchaeol.7 (2), 157–164.

alame, A., 1951. Le Syndrome Morgagni-Morel. Masson, Paris.assidy, A., Faughnan, M., 2000. Phyto-oestrogens through the life cycle. Proc. Nutr.

Soc. 59 (3), 489–496.aughey, J.E., 1958. The etiology of hyperostosis cranii (metabolic craniopathy); a

clinical study. J. Bone Joint Surg. Br. 40-B (4), 701–721.haljub, G., Johnson 3rd, R.F., Johnson Jr., R.F., Sitton, C.W., 1999. Unusually exuber-

ant hyperostosis frontalis interna: MRI. Neuroradiology 41 (1), 44–45.

laassen, H., 1989. Untersuchungen zur Anthropologie und Paläopathologie des

hallstattzeitlichen Menschen in der Oberpfalz. Thesis. Universität München.evriendt, W., Piercecchi-Marti, M.D., Adalian, P., Sanvoisin, A., Dutour, O., Leonetti,

G., 2005. Hyperostosis frontalis interna: forensic issues. J. Forensic Sci. 50 (1),143–146.

my 193 (2011) 453– 458

Fulton, J.D., Shand, J., Ritchie, D., McGhee, J., 1990. Hyperostosis frontalisinterna, acromegaly and hyperprolactinaemia. Postgrad. Med. J. 66 (771),16–19.

Gegick, C.G., Danowski, T.S., Khurana, R.C., Vidalon, C., Nolan, S., Stephan, T., Chae,S., Wingard, L., 1973. Hyperostosis frontalis interna and hyperphosphatasemia.Ann. Intern. Med. 79 (1), 71–75.

Glab, H., Szostek, K., Kaczanowski, K., 2006. Hyperostosisfrontalis interna, a geneticdisease?: Two medieval cases from Southern Poland. Homo. 57 (1), 19–27.

Hajdu, T., Fothi, E., Bernert, Z., Molnar, E., Lovasz, G., Kovari, I., Kohler, K., Marcsik, A.,2009. Appearance of hyperostosis frontalis interna in some osteoarcheologicalseries from Hungary. Homo 60 (3), 185–205.

Hall, R.C., Hall, R.C., 2003. Abuse of supraphysiologic doses of anabolic steroids.South. Med. J. 98 (5), 550–555.

Harding, F.E., 1949. Endocrinopathies associated with hyperostosis frontalis interna.Am. J. Med. 6 (3), 329–335.

Hasegawa, T., Ito, H., Yamamoto, S., Haba, K., Murata, H., 1983. Unilateral hyperos-tosis frontalis interna: case report. J. Neurosurg. 59 (4), 710–713.

Henschen, F., 1949. Morgani’s Syndrome. Oliver & Boyd, Edinburgh.Hershkovitz, I., Greenwald, C., Rothschild, B.M., Latimer, B., Dutour, O., Jellema, L.M.,

Wish-Baratz, S., 1999. Hyperostosis frontalis interna: an anthropological per-spective. Am. J. Phys. Anthropol. 109 (3), 303–325.

Lynnerup, N., Astrup, J.G., Sejrsen, B., 2005. Thickness of the human cranial diploë inrelation to age, sex and general body build. Head Face Med. 1, 13.

Moore, S., 1935. Hyperostosis frontalis interna. A preliminary study. Surg. Gynecol.Obstet. 61, 345–362.

Moore, S., 1936a. Calvarial hyperostosis and accompanying symptom-complex.Arch. Neurol. Psychiat. 35, 975.

Moore, S., 1936b. Metabolic craniopathy. Am. J. Roentgenol. 35, 30–39.Moore, S., 1955. Hyperostosis Cranii. CC Thomas, Springfield, 3–195.Morel, F., 1930. L’hyperostose frontale interne. Syndrome de l‘hyperostose frontale

interne avec adiposite et troubles cerebraux. Doin, Paris.Morgagni, G.B., 1719. Adversaria anatomica VI. Animadversio LXXIV. Vulporius,

Padua.Mulhern, D.M., Wilczak, C.A., Dudar, J.C., 2006. Brief communication: unusual finding

at pueblo bonito: multiple cases of hyperostosis frontalis interna. Am. J. Phys.Anthropol. 130 (4), 480–484.

Nikolic, S., Djonic, D., Zivkovic, V., Babic, D., Jukovic, F., Djuric, M., 2010. Rate ofoccurrence, gross appearance, and age relation of hyperostosis frontalis internain females: a prospective autopsy study. Am. J. Forensic Med. Pathol. 31 (3),205–207.

Pawlikowski, M., Komorowski, J.M., 1987. Hyperostosis frontalis and prolactin secre-tion. Exp. Clin. Endocrinol. 89 (1), 109–111.

Perou, M.L., 1964. Cranial Hyperostosis. Charles C. Thomas, Springfield, IL.Rudali, G., 1968. Experimental production of hyperostosis frontalis interna in mice.

Isr. J. Med. Sci. 4 (6), 1230–1235.Rühli, F.J., Boni, T., Henneberg, M., 2004. Hyperostosis frontalis interna: archaeolog-

ical evidence of possible microevolution of human sex steroids? Homo 55 (1/2),91–99.

Rühli, F.J., Henneberg, M., 2002. Are hyperostosis frontalis interna and leptin linked?A hypothetical approach about hormonal influence on human microevolution.Med. Hypotheses 58 (5), 378–381.

She, R., Szakacs, J., 2004. Hyperostosis frontalis interna: case report and review ofliterature. Ann. Clin. Lab. Sci. 34 (2), 206–208.

Smith, S., Hemphill, R.E., 1956. Hyperostosis frontalis interna. J. Neurol. Neurosurg.Psychiatry 19 (1), 42–45.

Sohmiya, M., Tanaka, M., Sotomatsu, A., Aihara, Y., Okamoto, K., 2001. Three elderlycases of hyperostosis cranii with various clinical symptoms. Nippon RonenIgakkai Zasshi 38 (2), 218–223.

Talarico Jr., E.F., Prather, A.D., Hardt, K.D., 2008. A case ofextensive hyperostosisfrontalis internain an 87-years-old female human cadaver. Clin. Anat. 21 (3),259–268.

Thevoz, F., 1966. The dura mater and its vessels in the morphogenesis of hyperostosisfrontalis interna. Ann. Anat. Pathol. (Paris) 11 (2), 121–149.

van Breda, E., Keizer, H.A., Kuipers, H., Wolffenbuttel, B.H., 2003. Androgenic anabolicsteroid use and severe hypothalamic-pituitary dysfunction: a case study. Int. J.Sports Med. 24 (3), 195–196.

Verdy, M., Guimond, J., Fauteux, P., Aube, M., 1978. Prevalence of hyperostosisfrontalis interna in relation to body weight. Am. J. Clin. Nutr. 31 (11), 2002–2004.

Yamakawa, K., Mizutani, K., Takahashi, M., Matsui, M., Mezaki, T., 2006. Hyperostosisfrontalis interna associated with hypogonadismin an elderly man. Age Ageing35 (2), 202–203.