Intracranial aneurysm as a possible complication of osteogenesis imperfecta: a case series and literature review

Abstract

Osteogenesis imperfecta (OI) is an inherited disease characterized by bone fragility due to impaired type I collagen. Although orthopedic management is improving, other complications are poorly understood. We describe three patients with OI with unruptured intracranial aneurysm (IA) detected by magnetic resonance angiography (MRA) screening of 14 patients. Case 1 was a 73-year-old woman with type 1 OI with blue sclera, vertebral compression fractures, and impaired hearing. Lumbar spine bone mineral density (BMD) was preserved (young adult mean (YAM): 86%). MRA revealed an IA in the right internal carotid artery. Case 2 was a 43-year-old man with type 4 OI and leg-length discrepancy due to left femoral neck fracture. Lumbar spine BMD was decreased (YAM: 61%). MRA showed an IA in the left anterior cerebral artery. Case 3 was a 35-year-old woman with type 3 OI with blue sclera, dentinogenesis imperfecta, deformity of the long bones, and severe scoliosis. She had undergone spine surgery and needed wheelchair assistance. The YAM of the femoral neck BMD was 71%. MRA indicated an IA in the right posterior communicating artery. The prevalence of IA in our series of patients with OI was 21%, which is higher than the reported prevalence of unruptured IA in the Japanese general population (2.2%), suggesting that IA may be a complication of OI. Our literature review revealed no cases of OI with unruptured IA, but 11 cases of OI with subarachnoid hemorrhage. IA seems unrelated to OI type, sex, or age. We recommend MRA of adults with OI.

OSTEOGENESIS IMPERFECTA (OI) is a genetic disorder characterized by recurrent fractures due to bone fragility. More than 90% of patients with OI have variants in either COL1A1 [MIM*120150] or COL1A2 [MIM*120160], which encode the α chains of collagen type I [1]. OI has various clinical presentations and is classified clinically as types 1 to 5 using the Sillence classification. Type 1 OI [MIM#166200] is a mild type with blue sclera and mild bone deformity and growth retardation. Type 2 OI [MIM#166210] is the most severe type that causes death due to multiple intracorporeal fractures, fractures during live birth, and respiratory failure. Type 3 OI [MIM#259420] is a severe type with progressive bone deformity, scoliosis, severe short stature, and characteristic facial features. Type 4 OI [MIM#166220] is a moderately severe type that cannot be classified as any of the other types, does not present as blue sclera, and has a mild degree of bone deformity and fracture susceptibility. Type 5 OI [MIM#610967] is a moderate type with interosseous calcification and hypertrophic callus.

Several recent studies have reported on patients with OI with subarachnoid hemorrhage (SAH) due to ruptured intracranial aneurysm (IA) [2-11]. However, there is no coherent study on unruptured IA as a complication of OI. Although patients with OI are constantly monitored by pediatricians during childhood, regular consultation with physicians is often discontinued in adulthood because the prevalence of fracture generally decreases. This situation increases the risk of signs of serious complications being overlooked. Herein, we report a case series of patients with OI with an unruptured IA found through magnetic resonance angiography (MRA) screening. We also discuss IA as a possible complication of OI and review the relevant literature.

Materials and Methods

We retrospectively searched the medical records of our institution to identify patients with OI who underwent head MRA. The patients had been diagnosed with OI based on clinical findings such as recurrent fractures, low bone mineral density (BMD), dentinogenesis imperfecta, blue sclera, and scoliosis, in accordance with the Japanese clinical practice guidelines [12]. The diagnoses were confirmed by genetic analyses. Head MRA was suggested for all 57 adult patients with OI. Fourteen of 57 patients agreed to undergo head MRA after considering the costs and risks. All 14 patients who underwent head MRA were included in this study. The mean age of these 14 patients was 41.0 ± 11.5 years, and the male to female ratio was 5:9 (Table 1). There was one patient with type 1 OI, six with type 4 OI, and seven with type 3 OI. Hypertension was defined as systolic and diastolic blood pressures of over 140 mmHg and 90 mmHg, respectively. Hypertension was detected in three of 14 patients; these three patients had no IA and did not take any antihypertensive medications.

Table 1 Baseline characteristics of our OI patients

Case No.	Age (years)	Sex	Clinical Type	Mutated Gene	YAM of BMD (%)	SBP/DBP (mmHg)	IA
1	73	female	1	COL1A1	86	116/66	+
2	43	male	4	COL1A2	61	128/78	+
3	35	female	3	unexamined	81*	97/60	+
4	50	male	4	COL1A1	83	137/83	–
5	47	female	3	unexamined	77	140/90**	–
6	46	female	4	COL1A1	79	119/71	–
7	44	female	3	COL1A2	73*	120/78	–
8	43	female	4	COL1A2	74	107/63	–
9	35	male	3	COL1A2	45	159/95**	–
10	34	male	4	COL1A1	96	125/69	–
11	33	female	3	COL1A1	36*	144/90**	–
12	31	male	4	COL1A1	76	110/68	–
13	30	female	3	COL1A1	36*	119/78	–
14	30	female	3	COL1A1	21	102/73	–

* These values were investigated by BMD of femoral neck.

** SBP >140 and DBP >90 were defined as hypertension.

OI, osteogenesis imperfecta; COL1A1/COL1A2, type 1 collagen alpha 1/2 chain;

YAM, young age mean; BMD, bone mineral density; SBP/DBP, systolic/diastolic blood pressure; IA, intracranial aneurysm

Written informed consent was obtained from all patients who underwent genetic analyses. Genetic analyses were performed in Okayama University and Kazusa DNA Research Institute using peripheral whole blood samples. Eleven out of 14 patients underwent genetic analysis as part of a clinical study at Okayama University using previously described methods [13]. In brief, genomic DNA sequencing for all exons of COL1A1, COL1A2, and IFTIM5 genes were determined by Sangar’s sequencing method [13]. Genetic analyses for 2 out of 14 patients were performed in Kazusa DNA Research Institute, an institution approved by the Japanese government, as described previously [14, 15]. In brief, DNA was purified from peripheral blood, and coding regions in exons and border regions of their introns of the listed genes were amplificated by polymerase chain reaction [14]. Genomic DNA sequences were determined by next-generation sequencing or capillary sequencing. The target genes were BMP1, COL1A1, COL1A2, CRTAP, FKBP10, IFITM5, P3H1, PP1B, SERPINF1, SERPINH1, SP7, TMEM38B, WNT1, CREB3L1, SPARC, TENT5A, MBTPS2, MESD, KDELR2, and CCDC134. Eight and four patients had variants in the COL1A1 and COL1A2 genes, respectively. Two patients did not wish to undergo genetic examination.

The BMD of the lumbar spine in each patient was examined using the dual-energy X-ray absorptiometry method (Discovery A: HOLOGIC Inc., Mississauga, ON, Canada). The BMD was evaluated by calculating the young adult mean (YAM) values, with low BMD defined as a YAM of <70%. The BMD of the femoral neck was measured in four patients who had undergone scoliosis surgery.

The literature review was conducted by searching PubMed (https://pubmed.ncbi.nlm.nih.gov/) using the key words “osteogenesis imperfecta” and “intracranial aneurysm.”

This study was approved by the Medical Ethic Committee of JCHO Osaka Hospital (ID: 2022-029). The clinical study for genetic analyses in Okayama University was approved by the Ethics Committee of Okayama University Hospital (No. 1701-038). All procedures performed in this study involving human participants were in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This study was an observational study by clinical chart review without any interventions and we disclosed the details of research information on our website, providing an opportunity for patients to opt-out.

Case 1

A 73-year-old woman with type 1 OI was diagnosed with a novel heterozygous variant in COL1A1 (p.Gly1052Cys). Her daughter and grandchildren had also been diagnosed with OI (Fig. 1a). She had a history of several traumatic fractures in her knees and toes after the age of 40–50 years, and two lumbar vertebral compression fractures. Her height was 143.0 cm (–2.84 SD), weight was 41.0 kg, and body mass index (BMI) was 20.0 kg/m². She had worn hearing aids since her 40s. She had blue sclera and slight scoliosis, but no obvious deformity of the lower extremities (Fig. 1b). The BMD of the lumbar spine was relatively preserved, with a YAM value of 86%. Head MRA revealed an unruptured IA (4.2 mm) in the right internal carotid artery (Fig. 1c).

Fig. 1

Skeletal and cerebrovascular manifestations in a patient with osteogenesis imperfecta type 1 (Case 1). a) Family tree showing that her daughter and grandchildren have inherited osteogenesis imperfecta. b) Skeletal radiographs show thick femoral cortical bone. c) Head magnetic resonance angiography shows an unruptured intracranial aneurysm (4.2 mm) in the right internal carotid artery (yellow arrow).

Case 2

A 43-year-old man with type 4 OI had been diagnosed with a variant in COL1A2 (p.Gly772Ser). His daughter was also diagnosed with OI (Fig. 2a). He had incurred over 10 fractures since childhood. His height was 161.2 cm (–1.65 SD), weight was 61.0 kg, and BMI was 23.5 kg/m². He had leg-length discrepancy due to left femoral neck fracture, but had no hearing loss or dentinogenesis imperfecta. Skeletal radiographs showed moderate scoliosis, mild deformity of the left tibia, an intramedullary nail in the left femur, and a screw in the right femoral neck (Fig. 2b). The BMD of the lumbar spine was decreased (YAM: 61%). Head MRA showed an unruptured IA (2.7 mm) in the left anterior cerebral artery (Fig. 2c).

Fig. 2

Skeletal and cerebrovascular manifestations in a patient with osteogenesis imperfecta type 4 (Case 2). a) Family tree showing that his daughter has inherited osteogenesis imperfecta. b) Skeletal radiographs show mild scoliosis, leg-length discrepancy (the right leg is 7.0 cm longer than the left), and an intramedullary nail in the left femur. c) Head magnetic resonance angiography shows an unruptured intracranial aneurysm (2.7 mm) in the left anterior cerebral artery (yellow arrow).

Case 3

A 35-year-old woman had been clinically diagnosed with type 3 OI because she did not wish to undergo genetic analysis. She was clinically diagnosed with OI at the age of 1.5 years due to multiple fractures. She had incurred innumerous fractures since childhood. She had no siblings and no recorded family history of recurrent fractures (Fig. 3a). Her height was 118.5 cm (–7.28 SD), weight was 29.5 kg, and BMI was 21.0 kg/m². She had blue sclera, dentinogenesis imperfecta, pigeon chest, and severe scoliosis and deformity of the long bones (Fig. 3b). She had undergone spinal fusion surgery, and needed wheelchair assistance. The BMD was slightly decreased in the left femoral neck (YAM: 71%) and left forearm (YAM: 70%). Head MRA showed an unruptured IA (3.0 mm) in the right posterior communicating artery (Fig. 3c).

Fig. 3

Skeletal and cerebrovascular manifestations in a patient with osteogenesis imperfecta type 3 (Case 3). a) Family tree showing no history of osteogenesis imperfecta. b) Skeletal radiographs show severe scoliosis, deformity of the long bones, and intramedullary nails in the spine and legs. c) Head magnetic resonance angiography shows an unruptured intracranial aneurysm (3.0 mm) at the confluence of the right posterior cerebral artery and posterior communicating artery (yellow arrow).

Discussion

In this report, we described the characteristics of three adults with OI and unruptured IA. The three with IA had no specific features to distinguish them from the total 14 patients with OI. As ruptured IA can be fatal, it is important to detect these lesions before they rupture.

We reviewed the literature to investigate the characteristics of patients with OI with IA. Although we found no reports describing unruptured IA in patients with OI, we found 11 cases of patients with OI with ruptured IA resulting in SAH [2-11] (Table 2). The mean age of the patients with OI with SAH was 43.3 ± 9.7 years, and the male to female ratio was 4:7. Five patients had type 1 OI and one had type 4 OI. One patient had a variant in COL1A1, and two had variants in COL1A2. Our three cases and the 11 previously reported cases suggest that OI can be complicated by IA, regardless of the clinical type or patient age. Type I collagen is expressed in the cerebral vascular wall and is thought to be involved in the strength and elasticity of the vascular wall [16]. In addition, a study of Japanese patients with IA showed that IAs are caused by single nucleotide polymorphisms in COL1A2 [17]. These studies suggest the possibility of IA as a complication of OI, implying that impaired type I collagen leads to fragility of the arterial walls.

Table 2 Literature review of 11 cases with ruptured aneurysms in adult OI patients

Authors	Age (years)	Sex	Clinical Type	Mutated Gene	Hypertension
Gaberel T et al. [2]	50	female	unknown	unknown	+
Gaberel T et al. [2]	50	female	unknown	unknown	+
Matouk CC et al. [3]	49	male	unknown	unknown	+
Okamura T et al. [4]	33	female	1	unknown	unknown
Narváez J et al. [5]	22	female	1	unknown	unknown
Havlik DM et al. [6]	38	male	4	unknown	unknown
Petruzzellis M et al. [7]	44	male	unknown	COL1A2	unknown
Kaliaperumal C et al. [8]	53	male	1	COL1A1	unknown
Hirohata T et al. [9]	37	female	1	not found	+
Mansfield K et al. [10]	50	female	1	COL1A2	–
Sardana V et al. [11]	50	female	unknown	unknown	unknown

OI, osteogenesis imperfecta; COL1A1/COL1A2, type 1 collagen α 1/2 chain

The reported prevalence of unruptured IA is 3.2% in the general population with a median age of 50 years [18] and 2.2% in the Japanese general population [19]. In our series, IA was detected in three of 14 adults with OI (21%) with a mean age of 50.3 ± 20.0 years, which suggests that the prevalence of IA may be higher in patients with OI than in the general population. Although our sample size was small, these data suggest that patients with OI may be more likely to have IA than the general population. Therefore, patients with OI should be carefully monitored by neurosurgeons.

A Japanese cohort study reported that the annual rupture rate of unruptured IA is 0.95% [20]. Furthermore, the reported risk factors for rupture are IA size, IA in the anterior or posterior communicating arteries, and irregular IA, while hypertension, smoking, multiple IA, and previous SAH are not risk factors for rupture. Another study reported that unruptured IA is found after the age of 30 years [19]. For patients with OI, who might have a higher prevalence of IA than the general population, we recommend they undergo head MRA during their 20s to 30s. In general, patients with unruptured IA are recommended to undergo follow-up head MRA every 6 to 12 months [21]. Because there may be a high prevalence of IA in patients with OI, and OI itself is a risk factor for rupture, we recommend head MRA and follow-up with neurosurgeons at least every 6 months regardless of the size, position, and shape of the IA. Even if no IA lesion is found, we suggest repeating the head MRA every 12 months to detect IA development.

In summary, we reported that three of 14 (21%) patients with OI had unruptured IA. Furthermore, ruptured IA resulting in SAH was previously reported in 11 patients with OI with various clinical types, ages, and mutated genes. Compared with the general population, the prevalence of IA may be higher in patients with OI. Thus, we recommend head MRA as a screening test for IA in adults with OI.

Acknowledgments

We thank our patients and their families for providing data through their medical records. We thank Dr. Kosei Hasegawa for performing the genetic analyses, and thank Dr. Yuiko Nagamatsu, Dr. Kanako Kishimoto, Dr. Kawai Kondo, and Dr. Natsuko Sakamoto for their advice. We also thank the clinical staff of Osaka Hospital, JCHO. We thank Dr. Kelly Zammit, BVSc, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.

Disclosure

None of the authors have any potential conflicts of interest associated with this research.

References

• 1   Rauch  F,  Glorieux  FH (2004) Osteogenesis imperfecta. Lancet 363: 1377–1385.
• 2   Gaberel  T,  Rochey  A,  di Palma  C,  Lucas  F,  Touze  E, et al. (2016) Ruptured intracranial aneurysm in patients with osteogenesis imperfecta: 2 familial cases and a systematic review of the literature. Neurochirurgie 62: 317–320.
• 3   Matouk  CC,  Hanbidge  A,  Mandell  DM,  Terbrugge  KG,  Agid  R (2011) Osteogenesis imperfecta, multiple intra-abdominal arterial dissections and a ruptured dissecting-type intracranial aneurysm. Interv Neuroradiol 17: 371–375.
• 4   Okamura  T,  Yamamoto  M,  Ohta  K,  Matsuoka  T,  Takahashi  M, et al. (1995) A case of ruptured cerebral aneurysm associated with fenestrated vertebral artery in osteogenesis imperfecta. No Shinkei Geka 23: 451–455 (In Japanese).
• 5   Narváez  J,  Narváez  JA,  Majós  C,  Clavaguera  MT,  Alegre-Sancho  JJ (1996) Subarachnoid haemorrhage secondary to ruptured cerebral aneurysm in a patient with osteogenesis imperfecta. Br J Rheumatol 35: 1332–1333.
• 6   Havlik  DM,  Nashelsky  MB (2006) Ruptured cerebral artery aneurysm and bacterial meningitis in a man with osteogenesis imperfecta. Am J Forensic Med Pathol 27: 117–120.
• 7   Petruzzellis  M,  De Blasi  R,  Lucivero  V,  Sancilio  M,  Prontera  M, et al. (2007) Cerebral aneurysms in a patient with osteogenesis imperfecta and exon 28 polymorphism of COL1A2. AJNR Am J Neuroradiol 28: 397–398.
• 8   Kaliaperumal  C,  Walsh  T,  Balasubramanian  C,  Wyse  G,  Fanning  N, et al. (2011) Osteogenesis imperfecta presenting as aneurysmal subarachnoid haemorrhage in a 53-year-old man. BMJ Case Rep 2011: bcr1020114910.
• 9   Hirohata  T,  Miyawaki  S,  Mizutani  A,  Iwakami  T,  Yamada  S, et al. (2014) Subarachnoid hemorrhage secondary to a ruptured middle cerebral aneurysm in a patient with osteogenesis imperfecta: a case report. BMC Neurol 14: 150.
• 10   Mansfield  K,  Rahme  R (2015) Dissecting aneurysm of the recurrent artery of Heubner in a patient with osteogenesis imperfecta. Can J Neurol Sci 42: 461–465.
• 11   Sardana  V,  Kamble  S,  Sharma  SK,  Maheshwari  D,  Bhushan  B (2017) Mirror aneurysm with right frontal ICH in a patient with osteogenesis imperfecta. J Assoc Physicians India 65: 103–105.
• 12  (2018) Clinical guideline for osteogenesis imperfecta. In: The Japanese Society for Pediatric Endocrinology (ed) Clinical Practice Guidelines. Nakayama Shoten, Tokyo, Japan: 228–236 (In Japanese).
• 13   Higuchi  Y,  Hasegawa  K,  Futagawa  N,  Yamashita  M,  Tanaka  H, et al. (2021) Genetic analysis in Japanese patients with osteogenesis imperfecta: genotype and phenotype spectra in 96 probands. Mol Genet Genomic Med 9: e1675.
• 14   Fujiki  R,  Ikeda  M,  Yoshida  A,  Akiko  M,  Yao  Y, et al. (2018) Assessing the accuracy of variant detection in cost-effective gene panel testing by next-generation sequencing. J Mol Diagn 20: 572–582.
• 15   Fujiki  R,  Ikeda  M,  Ohara  O (2019) Short DNA probes developed for sample tracking and quality assurance in gene panel testing. J Mol Diagn 21: 1079–1094.
• 16   Prockop  DJ,  Kivirikko  KI (1995) Collagens: molecular biology, diseases, and potentials for therapy. Annu Rev Biochem 64: 403–434.
• 17   Yoneyama  T,  Kasuya  H,  Onda  H,  Akagawa  H,  Hashiguchi  K, et al. (2004) Collagen type I α2 (COL1A2) is the susceptible gene for intracranial aneurysms. Stroke 35: 443–448.
• 18   Vlak  MH,  Algra  A,  Brandenburg  R,  Rinkel  GJ (2011) Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis. Lancet Neurol 10: 626–636.
• 19   Iwamoto  H,  Kiyohara  Y,  Fujishima  M,  Kato  I,  Nakayama  K, et al. (1999) Prevalence of intracranial saccular aneurysms in a Japanese community based on a consecutive autopsy series during a 30-year observation period. Stroke 30: 1390–1395.
• 20   Morita  K,  Kirino  T,  Hashi  K,  Aoki  N,  Fukuhara  S, et al. (2012) The natural course of unruptured cerebral aneurysms in a Japanese cohort. N Engl J Med 366: 2474–2482.
• 21  The Japan Stroke Society (ed) (2021) Japanese guideline for the management of stroke 2021. Kyowa Kikaku, Tokyo, Japan (In Japanese).