Journal of Human Hypertension (2000) 14, 91–94  2000 Macmillan Publishers Ltd All rights reserved 0950-9240/00 $15.00 www.nature.com/jhh ORIGINAL ARTICLE Alpha-1-antitrypsin phenotypes in patients with renal arterial fibromuscular dysplasia

A Bofinger, C Hawley, P Fisher, N Daunt, M Stowasser and R Gordon Hypertension Unit, University of Queensland Department of Medicine, Greenslopes Private Hospital, Brisbane, Queensland, Australia

␣ Fibromuscular dysplasia (FMD) is a significant cause of one (1.2%) PiSZ), suggesting that 1-AT deficiency is renal artery stenosis, especially in young females. A not a common aetiological factor in renal arterial FMD. ␣ ␣ rare association between FMD and 1-antitrypsin ( 1- However, despite FMD being three times less common ␣ AT) deficiency has been reported. We compared the 1- in males than females, and carotid artery dissection AT phenotype distribution in 83 patients with renal being a rare occurrence, a male with PiMS deficiency arterial FMD with those published for Australian popu- phenotype presented with internal carotid artery dissec- ␣ lations. 1-AT phenotyping was performed by isoelectric tion and had bilateral renal artery FMD. Further, a patient focusing between pH 4.2 and pH 4.9 on polyacrylamide with PiSZ deficiency phenotype was one of two sisters gels with PiM1M2, PiFM (non-deficiency alleles), PiMS with FMD and was more severely affected than her PiMM and PiMZ (deficiency alleles) markers. Following pheno- normal phenotype sibling. These two patients from the ␣ typing, 1-AT genotyping was performed in 10 patients present series together with nine culled from the litera- ␣ to confirm the presence of S and/or Z alleles. The ture with 1-AT deficiency phenotype and FMD suggest ␣ phenotype distribution and allele frequencies were simi- that the chance combination of 1-AT deficiency and lar to those reported for normal subjects from two Aus- FMD may predispose to severe manifestations of FMD. tralian populations (72 (86.7%) PiMM phenotype, one Journal of Human Hypertension (2000) 14, 91–94 (1.2%) PiFM, seven (8.4%) PiMS, two (2.4%) PiMZ and

␣ Keywords: fibromuscular dysplasia; 1-antitrypsin deficiency; renovascular hypertension; renal artery; internal carotid artery; aetiology

␣ Introduction While an association between FMD and 1-AT deficiency has been reported,3–9 we are aware of Fibromuscular dysplasia (FMD) is a non-atheroscler- only nine patients in all. Other rare associations otic, non-inflammatory vascular disease1 most com- ␣ between 1-AT deficiency and other cardiovascular monly affecting the renal, internal carotid and exter- such as intracranial aneurysms,10,11 cer- nal iliac arteries. FMD is the most common cause ebral artery dissection,8,10 mesenteric artery aneur- of renovascular hypertension in patients presenting ysm rupture,12 and iliac artery dissection13 have also under 40 years of age, and affects women three times 1 been reported. more frequently than men. Although rare, arterial In the current study we tested the hypothesis that dissection is a complication of FMD, and is recog- patients with renal arterial FMD have a higher nised as a cause of stroke in the young.1 ␣ prevalence of 1-AT deficiency than the general The most abundant proteinase inhibitor in human ␣ ␣ ␣ community. We determined 1-AT phenotypes in 83 plasma is 1-antitrypsin ( 1-AT) and is thought to be patients with renal arterial FMD, and compared important in protection against , parti- 2 these to published phenotype distributions in Aus- cularly neutrophil elastase. The highly polymor- 14,15 ␣ tralian populations. phic 1-AT identified at C14q32 has more than 70 allelic variants, most of which do not result in ␣ 2 Materials and methods decreased plasma 1-AT levels. The alleles most ␣ commonly associated with 1-AT deficiency are S One hundred and seven patients referred to a hyper- 2 ␣ and Z. Homozygotes for the Z allele have 1-AT lev- tension unit between 1970 and 1997 were diagnosed els which are only 12–18% of normal, and are at with FMD of the renal arteries. Twenty-four patients increased risk of lung and disease.2 The PiMZ were excluded from the present study, comprising ␣ and PiMS phenotypes have mean 1-AT levels eight who were deceased, 11 who were lost to fol- approximately 65% and 85% of normal, respect- low-up, and five from whom an appropriate blood ively.2 sample could not readily be obtained. Causes of death in the eight deceased patients were (four patients), alcohol-related liver disease (one Correspondence: Professor Richard D Gordon, University Depart- ment of Medicine, Greenslopes Private Hospital, Newdegate patient), postoperative haemorrhage (one patient), Street, Brisbane Q 4120, Australia myocardial infarction (one patient) and unknown Received 7 June 1999; revised and accepted 1 September 1999 (one patient). Alpha-1-antitrypsin deficiency and FMD A Bofinger et al 92 The remaining 83 patients were interviewed by (PiMS, PiMZ or PiSZ). Allele frequencies for M, F, one of the authors (AB). Duration of hypertension S and Z were 0.927, 0.006, 0.048 and 0.018 respect- was determined as the number of years between the ively. These are similar to those reported for Aus- first diagnosis of sustained hypertension and the tralian normals14,15 and also for several different diagnosis of FMD. Caucasian populations.2,17,18 The diagnosis of FMD was confirmed by angio- Comparison of the 10 deficiency phenotype graphic (80 patients) and/or pathologic (21 patients) patients with the 73 non-deficiency phenotype examination. All angiograms were reviewed by three patients revealed no significant differences in gen- of the authors (AB, CH, ND). Each FMD-affected der distribution, age at diagnosis of FMD, duration renal artery was classified as unifocal (focal or of hypertension, type of FMD lesion, number with tubular) or multifocal. A focal lesion was defined as bilateral FMD, nor number with unilateral renal a solitary stenosis of less than 1 cm in length, and atrophy. a tubular lesion as a smooth, concentric stenosis of However, two patients with deficiency pheno- greater than 1 cm in length. A multifocal lesion was types are of interest. A 50-year-old male with PiMS defined as two or more adjacent stenoses deficiency phenotype suffered a spontaneous dissec- (significantly more in the majority) giving the tion of the right internal carotid artery and was characteristic ‘beaded’ appearance. Lesions were found to have bilateral renal arterial FMD. The sin- classified as unilateral (right or left) or bilateral. gle patient with PiSZ phenotype in this series was ‘Unilateral renal atrophy’ was defined as a differ- one of two sisters with FMD, and presented aged 44 ence in bipolar length of greater than 2 cm between with bilateral renal artery and iliac artery multifocal the two kidneys on IVP or renal angiography. FMD, severe right renal artery stenosis and unilat- ␣ 1-AT phenotyping was performed by isoelectric eral renal atrophy. She had no clinical evidence of focusing between pH 4.2 and pH 4.9 on polyacryla- lung or liver pathology. Her sister with FMD and mide gels using PiM1M2, PiMS, PiMZ and PiFM PiMM phenotype presented at age 57 with bilateral ␣ markers. Following phenotyping, 1-AT genotyping mild renal artery FMD, no renal atrophy and iliac was performed in 10 patients, using the method of artery multifocal FMD. Anderson et al16 to confirm the presence of S and/or Z alleles. Discussion Statistical analysis was performed using Statistica software (Release 5.0, StatSoft, Inc. Tulsa, USA). In a retrospective review of 6696 autopsies perfor- Parametric data were expressed as mean (± standard med at the Mayo Clinic,3 25 patients had FMD, six ␣ deviation), and non-parametric as median (range). had 1-AT deficiency and only two (with PiZZ The t-test for independent samples was used for phenotype) had both. In one of these, the FMD affec- parametric data, the Mann–Whitney U test for non- ted the renal arteries, and, in the other, the iliac ␣ parametric data, the Fisher exact test for differences arteries. Both patients with FMD and 1-AT between proportions, and the Chi square test for dif- deficiency died of disseminated fungal infection ferences between observed and expected phenotype with mycotic aneurysms during post-transplant frequencies. A two-sided probability value of less immunosuppressive therapy. than 0.05 was considered statistically significant. A search of the literature revealed only seven ␣ additional patients with FMD and 1-AT deficiency.4–9 Only one of these was said to have Results 9 ␣ renal arterial FMD. A 15-year-old girl with 1-AT Sixty patients (54 female) had multifocal disease deficiency (PiZZ phenotype) died of a ruptured and 23 (17 female) had unifocal disease. Seventy- splenic artery aneurysm associated with medial fib- two (86.7%) of the patients were PiMM phenotype, roplasia following orthotopic liver transplantation.4 ␣ one (1.2%) PiFM, seven (8.4%) PiMS, two (2.4%) A 16-year-old girl with mild 1-AT deficiency (PiMP PiMZ and one (1.2%) PiSZ (Table 1). That is, 10 phenotype) had a giant intracranial aneurysm of the (12%) of the patients had deficiency phenotypes middle cerebral artery associated with intimal 5 ␣ FMD. An 11-year-old boy with 1-AT deficiency ␣ (PiSZ phenotype) had a left middle cerebral artery Table 1 Observed and expected 1-AT phenotype distribution in 83 FMD patients occlusion and a ‘string-of-beads’ appearance in the supraclinoid portion of the left internal carotid 6 ␣ FMD patients Expected artery. Two females with 1-AT deficiency (both distributiona PiMZ phenotype) had bilateral internal carotid sten- 7 ␣ Multifocal Unifocal Total osis caused by FMD. A 59-year-old female with 1- AT deficiency (phenotype PiMS) had a dissecting Phenotype aneurysm of the right vertebral artery, a saccular MM 51 21 72 73 aneurysm of the posterior communicating artery, FM 1 0 1 0 8 b and FMD of the left vertebral artery. A 47-year-old MS 5278 ␣ MZb 2022female with 1-AT deficiency (PiZZ phenotype) and SZb 1010mild pulmonary symptoms presented with rupture ZZb 0000of the right renal artery 25 years after removal of a Total 60 23 83 83 small, non-functioning left kidney.9 The present study of 83 patients with renal a b ␣ Derived from references (14,15); deficiency phenotypes. arterial FMD found a similar distribution of 1-AT

Journal of Human Hypertension Alpha-1-antitrypsin deficiency and FMD A Bofinger et al 93 deficiency phenotypes to that in the general popu- there is a possibility that the chance association of ␣ ␣ lation, suggesting that an 1-AT deficiency pheno- FMD and 1-AT deficiency led to more severe mani- type is not a common aetiological factor in the festations of FMD. development of renal arterial FMD. The literature would support this, since such an association has Acknowledgements been rarely reported. However, in two of 10 patients ␣ with 1-AT deficiency phenotype in the present ser- This work was supported by grants from the Aus- ␣ ies, it is possible that the 1-AT deficiency contrib- tralian Kidney Foundation and the Sir Edward Dun- uted to the severity of the disease. Thus, in two sis- lop Medical Research Foundation. AB receives a ters with FMD, the sister with the deficiency (PiSZ) Medical Postgraduate Research Scholarship from phenotype presented at a younger age and with a the National Health and Medical Research Council significantly more severe degree of disease than her of Australia. MS is a Postdoctoral Fellow with the PiMM phenotype sibling. A second patient with a National Heart Foundation of Australia. deficiency phenotype (PiMS) and internal carotid artery dissection and bilateral renal artery FMD was References of male gender, where the incidence of FMD is lower, and internal carotid artery dissection is a rare 1Lu¨ scher TF et al. Arterial fibromuscular dysplasia. manifestation of FMD in either sex.1 Mayo Clin Proc 1987; 62: 931–952. ␣ Elastase is capable of degrading a variety of sub- 2 Cox DW. 1-Antitrypsin deficiency. In: Scriver CR, strates,2 including elastin, types III and IV, Beaudet AL, Sly WS, Valle D (eds). The Metabolic and 19–21 ␣ Molecular Bases of Inherited Disease. 7th edn. fibronectin and the , and 1- McGraw-Hill: New York, 1995, pp 4125–4158. AT, which inactivates a wide range of proteases, 3 Schievink WI, Bjornsson J, Parisi JE, Prakash UB. provides approximately 90% of the antielastase Arterial fibromuscular dysplasia associated with sev- activity in plasma.2 Excessive elastase activity is the ere alpha 1-antitrypsin deficiency. Mayo Clin Proc ␣ accepted basis of lung damage in 1-AT deficiency 1994; 69: 1040–1043. states in humans: destruction of elastic fibres by 4 Brems JJ et al. Splenic artery aneurysm rupture follow- human neutrophil elastase produces emphysema in ing orthotopic liver transplantation. Transplantation experimental animal models.22 It has been suggested 1988; 45: 1136–1137. ␣ 5 Schievink WI et al. Giant intracranial aneurysm and that 1-AT may also have an important role in pro- tecting blood vessels.2,23 When temporal arteries fibromuscular dysplasia in an adolescent with alpha 1- antitrypsin deficiency. J Neurosurg 1996; 85: 503–506. from 24 patients with intracranial aneurysms were 6So¨lder B et al. Fibromuscular dysplasia of the internal compared with those from 10 patients with benign, carotid artery in a child with alpha-1-antitrypsin non-vascular intracranial pathology, the group with deficiency. Dev Med Child Neurol 1997; 39: 827–829. aneurysms had higher plasma elastase levels and 7 Schievink WI, Meyer FB, Parisi JE, Wijdicks EF. Fibro- immunohistological evidence of greater degradation muscular dysplasia of the internal carotid artery asso- of arterial elastin.24 In contrast with a 2–4% preva- ciated with alpha l-antitrypsin deficiency. Neurosur- lence in the general population, 11% of 47 patients gery 1998; 43: 229–233. ␣ 8 Schievink WI, Katzmann JA, Piepgras DG. Alpha-1- with abdominal aortic aneurysm had an 1-AT 25 ␣ antitrypsin deficiency in spontaneous intracranial deficiency phenotype (PiMZ). A deficiency of 1- AT could therefore result in arterial wall degra- arterial dissections. Cerebrovasc Dis 1998; 8: 42–44. 9 Randers E, Jonler M, Lund AM, Danielsen H. Atraum- dation through inadequate protection against pro- 2,7,23–25 atic loss of a kidney in a patient with alpha-l-antitryp- teolytic activity. An effect such as this might sin deficiency. Nephron 1998; 80: 71–75. predispose to increased severity of multifocal FMD 10 Schievink WI, Prakash UB, Piepgras DG, Mokri B. and also to arterial dissection in FMD. Alpha 1-antitrypsin deficiency in intracranial aneur- While the prevalence of renal arterial FMD in ysms and cervical artery dissection. Lancet 1994; 343: autopsy studies has been only approximately 1%,3,26 452–453. renal artery FMD has been reported in 3.5% of 11 Schievink WI, Katzmann JA, Piepgras DG, Schaid DJ. potential living renal transplant donors,27 who Alpha-1-antitrypsin phenotypes among patients with would have been carefully studied, bringing into intracranial aneurysms. J Neurosurg 1996; 84: 781– question the true prevalence of FMD. Similarly, 784. ␣ 12 Mitchell MB, McAnena OJ, Rutherford RB. Ruptured while severe 1-AT deficiency (PiZZ phenotype) is mesenteric artery aneurysm in a patient with alpha l- considered rare, partial deficiency (PiMS and PiMZ antitrypsin deficiency: etiologic implications. J Vasc phenotypes) has been described in 10% to 15% of Surg 1993; 17: 420–424. the Australian population14,15 and was present in 13 Cattan S, Mariette X, Labrousse F, Brouet JC. Iliac 12% of patients in the present study. artery dissection in alpha 1-antitrypsin antitrypsin Since milder forms of ␣ -AT deficiency are not deficiency. Lancet 1994; 343: 1371–1372. 1 ␣ uncommon, and the prevalence of FMD is uncertain, 14 Mulley JC. Distribution of the six 1-antitrypsin (PI) chance associations between the two might be subtypes from a sample of blood donors. Aust J Exp ␣ Biol Med Sci 1980; 58: 409–414. expected. Since the frequency of 1-AT deficiency was no higher in the patients with FMD in the 15 Clark P. Alpha-1- inhibitor phenotypes in Australia. Hum Hered 1982; 32: 225–227. present study than that reported for the general ␣ 16 Anderson BS et al. Two novel nonradioactive poly- population, it is very unlikely that 1-AT deficiency merase chain reaction based assays of dried blood is a common aetiological factor in the development spots, genomic DNA, or whole cells for fast, reliable of FMD. However, in two individual patients in the detection of Z and S alleles in the alpha-1-antitrypsin present series and nine culled from the literature, gene. Clin Chem 1992; 38: 2100–2107.

Journal of Human Hypertension Alpha-1-antitrypsin deficiency and FMD A Bofinger et al

94 ␣ 17 Dykes DD, Miller SA, Polesky HF. Distribution of 1- 22 Snider GL, Lucey EC, Stone PJ. Animal models of antitrypsin phenotypes in a US white population. emphysema. Am Rev Respir Dis 1986; 133: 149–169. Hum Hered 1984; 34: 308–310. 23 Cox DW. Alpha 1-antitrypsin: a guardian of vascular 18 DeCroo S, Kamboh MI, Ferrell RE. Population genetics . Mayo Clin Proc 1994; 69: 1123–1124. of alpha-1-antitrypsin polymorphism in US whites. 24 Connolly ES, Jr. et al. Elastin degradation in the super- Hum Hered 1991; 41: 215–221. ficial temporal arteries of patients with intracranial 19 Mainardi CL, Dixit SN, Kang AH. Degradation of type aneurysms reflects changes in plasma elastase. Neuro- IV (basement membrane) collagen by a proteinase iso- surgery 1997; 40: 903–909. lated from human polymorphonuclear leukocyte gran- 25 Cohen JR, Sarfati I, Ratner L, Tilson D. Alpha 1-anti- ules. J Biol Chem 1980; 255: 5435–5441. phenotypes in patients with abdominal aortic 20 McDonald JA, Kelley DG. Degradation of fibronectin aneurysms. J Surg Res 1990; 49: 319–321. by human leukocyte elastase. Release of biologically 26 Heffelfinger MJ, Holley KE, Harrison EG, Jr, Hunt JC. active fragments. J Biol Chem 1980; 255: 8848–8858. Arterial fibromuscular dysplasia studied at autopsy 21 Mainardi CL, Hasty DL, Seyer JM, Kang AH. Specific (abstract). Am J Clin Pathol 1970; 54: 274. cleavage of human type III collagen by human poly- 27 Cragg AH et al. Incidental fibromuscular dysplasia in morphonuclear leukocyte elastase. J Biol Chem 1980; potential renal donors: long-term clinical follow-up. 255: 12006–12010. Radiology 1989; 172: 145–147.

Journal of Human Hypertension