Prevalence and Prevention of Avascular Necrosis of Bone in SLE

1
Prevalence and Prevention of Avascular Necrosis
of Bone in SLE   H. Michael Belmont,
M.D. Director, Bellevue Hospital Lupus
Clinic Chief Medical Officer, Hospital for Joint
Diseases Associate Professor of Medicine New York
University School of Medicine
2
Etiology of Bone Infarct/AVN/Osteonecrosis
Traumatic
Idiopathic (e.g. Legg-Calve-Perthes, Kienbocks,
etc.)
Atraumatic Vasculopathy Fat embolism/Adipose
expansion
3

• Systemic lupus erythematosus (SLE) is
  characterized by immune dysregulation that
  results in the production of autoantibodies,
  generation of circulating immune complexes, and
  activation of the complement system.
• A pathological hallmark of SLE is the recurrence
  of widespread and diverse vascular lesions
  inflammatory and thrombotic.
• The endothelial injury typical of SLE flares may
  serve as an inciting event, which predisposes to
  the accelerated atherosclerosis that is
  associated with the disorder as well as disrupt
  the microcirculation (e.g. end artery of the
  femoral head) to predispose to AVN.

PATHOBIOLOGY OF SLE
4
As a result, SLE patients exhibit rates of
myocardial infarction and cerebrovascular
accident that are up to 50-fold higher than age
and gender matched controls. The risk of
developing avascular necrosis is 10-40 times
greater than other patients on corticosteroids. T
he increased frequency of coronary artery disease
(CAD) and AVN observed in patients with SLE may
be unified by the underlying vascular injury that
distinguishes the disease.
5
Pathologic and Clinical Spectrum of Vasculopathy
in SLE

• Pathology Pathogenesis Clinical
  Phenomenon
• Capillaritis Immune complex deposition Glomerulo
  nehpritis, pulmonary alveolar
• Vasculitis Activation of complement,
  hemorrhage
• neutrophils, and endothelium Cutaneous
  purpura, polyarteritis nodosa-like
• Modeled by Arthus lesion
  systemic and cerebral vasculitis, AVN
• Leukothrombosis Intravascular activation of
  complement, Widespread vascular
  injury, hypoxia, cerebral
• neutrophils, and vascular
  endothelium cerebral
  dysfunction, SIRS, AVN
• Absence of local immune complex
• deposition
• Modeled by Shwartzman lesion
• Thrombosis Antibodies to anionic
  phospholipid-protein Arterial and
  venous thrombosis, fetal wastage,
• complexes interact with endothelial cells,
  thrombocytopenia, pulmonary
  hypertension, AVN
• platelets, or coagulation factors
• Modeled by APS
• Disseminated intravascular
  platelet TTP
• aggregation

6
Evidence for Shwartzman Phenomenon in
SLE   Increased C3a, C5a Increased neutrophil
CD11b/CD18 (beta 2 integrin, CR3) Increased
endothelial cell adhesion molecules Increased
endothelial cell nitric oxide synthase Increased
circulating endothelial cells Histologic
evidence of leukoaggregates (CNS, mesentery)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
Endothelial Cell Adhesion Molecule Expressionin
Active versus Inactive SLE
P lt0.01 active vs. control P lt0.025 active
vs. inactive


Immunohistochemical score

T
Belmont, Buyon, Giorno, Abramson Arthritis
Rheum, 1994
12
(No Transcript)
13
Increased NO Accompanied by the Upregulation of
iNOS in Vascular Endothelium of SLE Patients
P lt.05 active vs. inactive
P lt.01 active vs. inactive
Immunohistochemical score
Belmont et al., Arthritis Rheum, 1997
14
Confirmation of Increased CEC in Peripheral
Blood from Patients with Active SLE (
Positive Selection on P1H12 Magnetic Beads)
CEC/ml
N 8
N 15
N 9
15
(No Transcript)
16
SLE and Bone Infarct/AVN

• AVN is a frequent morbidity in lupus.
• The incidence is estimated to be between 5-80.
• SLE patients have the highest rate of AVN as
  compared to rheumatoid arthritis, pemphigoid, and
  asthma. 

17
Prevalence of AVN in SLE

• Two prospective studies of high dose
  corticosteroid treated SLE patients report a
  prevalence of AVN between 19-80
• Aranow J. Rheumatology 199724
• Nagasawa British Journal of Rheumatology
  199433
• Aranow screened 62 SLE patients by MRI for AVN.
  43 of 62 patients took ? 30mg/day of prednisone
  and 9 (19) had evidence of AVN. Patients who
  had taken lt 30mg/day had no evidence of AVN.
• Nagasawa studied 23 SLE patients treated with
  high dose corticosteroids (? 20mg/day), 8 (30)
  had evidence of AVN by MRI. This cohort was
  followed prospectively for 3 years. Five of
  these patients received ? 30mg/day and 4 (80)
  newly developed AVN

18
Risk Factors for AVN in SLE

• Daily steroid dosage
• Cushingoid body habitus
• Age
• Ethnicity
• Vasculitis
• Raynauds
• Thrombophlebitis
• Presence of cardiolipin antibody
• Increased disease activity
• Bergstein, J Paediatrics 17485
• Velayos, Ann Intern Med 196664
• Klipper Medicine 1976 55
• Zizic Am J Med 1985 79

19
Novel Paradigm

• We propose the high prevalence and multifocal
  nature of AVN in SLE patients during high dose
  steroid treatment of disease exacerbation results
  from the co-occurrence of two processes
• 1. Vascular Injury with widespread activation
  of endothelium and inflammatory vasculopathy or
  thrombosis disrupting the microcirculation
• 2. Lipid Deposition with interosseous fat
  accumulation, secondary to glucocorticoid induced
  abnormal lipid metabolism, increasing
  intramedullary pressure (i.e. bony compartment
  syndrome) and decreasing perfusion

20
Materials and Methods

• The ICD-9 code for SLE, 710.0, and for AVN
  733.40-44, 49 were cross-matched in a computer
  search of the medical records departments of
  Bellevue Hospital, Hospital for Joint Diseases
  (HJD), and the HJD faculty practice for the
  18-month period between 08/99 and 02/00.
• In addition 170 charts were reviewed of patients
  who met ACR criteria for SLE
• 44 charts of patients with AVN and SLE were used
  to generate the case report forms.

21
RESULTS

• 44 patients, 42 female, 2 male and mean age 38
  (19-61)
• 18 Hispanic, 13 African American, 12 Caucasian, 1
  Asian
• 33 ascertained by MRI and 11 established at joint
  replacement
• 42 ANA (), 36 dsDNA(), 22 nephritis, 4
  cerebritis, 10 ACA(), 5 history of thrombotic
  event
• 18 Hyperlipidemia, 1CAD, 24 HTN, 5
  Postmenospausal, 1 Diabetes Mellitus, 4 cigarette
  smoking, 5 Family History CAD, mean cholesterol
  233/-14 mg/dl

22
RESULTS

• 44 patients, 42 female, 2 male and mean age 38
  (19-61)
• 18 Hispanic, 13 African American, 12 Caucasian, 1
  Asian
• 33 ascertained by MRI and 11 established at joint
  replacement
• 42 ANA (), 36 dsDNA(), 22 nephritis, 4
  cerebritis, 10 ACA(), 5 history of thrombotic
  event
• 18 Hyperlipidemia, 1CAD, 24 HTN, 5
  Postmenospausal, 1 Diabetes Mellitus, 4 cigarette
  smoking, 5 Family History CAD, mean cholesterol
  233/-14

23
RESULTS

• 5 Unifocal AVN, 25 bifocal AVN, with 24/25
  bilateral femoral head, 15 multifocal with 3 or
  more joints affected
• 95 received prednisone ? 60 mg/day and 97
  received ? 30 mg/day. 97 received high dose
  prednisone for a SLEDAI ? 8

24
SUMMARY

• AVN in SLE is unifocal in only 9 (5/44)
• AVN in SLE affects more than one joint in 82
  (39/44)
• AVN in SLE affects bilateral femoral head in 50
  (24/44)
• AVN in SLE is multifocal with ? 3 joints in 34
  (15/44)
• AVN in SLE is associated with high dose steroid
  therapy for increased disease activity

25
CONCLUSION

• The finding that AVN in steroid treated SLE
  patients is most often multifocal is consistent
  with our hypothesis that diffuse vascular injury
  during disease activity conspires with steroid
  induced marrow lipocyte accumulation to compress
  blood vessels and results in ischemic bone
  necrosis. Clinical trials with statins, which by
  preventing abnormal fat metabolism and
  endothelial injury may reduce this morbidity, are
  warranted.

26
Increased Adipogenesis and Abnormal Lipid
Metabolism is Associated with Corticosteroid-induc
ed Osteonecrosis of Bone in Animal Models

• Clofibrate, decreased the corticosteroid-induced
  changes on marrow fat conversion and the increase
  in femoral head pressure
• In the group only treated with methylprednisilone,
  the percentage of fatty marrow was increased by
  an average of 28, and the average intrafemoral
  head pressure increased from a baseline of 25 to
  55.
• In the group treated with steroids and clofibrate
  there was almost no change in the percentage of
  fatty marrow and there was no increase in the
  intrafemoral head pressure.

27

• Cui showed that a lipid-clearing agent prevented
  osteonecrosis in a chicken model
• 25 chickens were injected with methylprednisolone
  and 10 received methylprednisolone plus
  lovastatin.
• At autopsy 56 or 14/25 of the chickens only
  given corticosteroids had some evidence of
  necrosis, with 4/25 demonstrating evidence of
  sub-chondral death and resorption and new bone
  formation.
• In the group given lovastatin, there was a
  smaller conversion to fatty marrow and 10/10 had
  no evidence of osteonecrosis.
• In the study, the authors were able to show that
  lovostatin counteracted the effect of steroids on
  the differentiation of precursor cells in bone
  marrow into adipocytes.

28

• 14 of 20 rabbits with induced hypersensitivity
  vasculitis and high dose corticosteroid exposure
  developed evidence of AVN in the femoral
  metaphysis.
• In the groups only given corticosteroids without
  induction of hypersensitivity vasculitis, there
  was no evidence of AVN.
• Animal experiments done by Matsui, showed a
  possible synergy between vasculitis, high-dose
  cortocosteroids and AVN.
• Motomura Steroid induced osteonecrosis on rabbits
  No rx 14/20 (70), warfarin 7/21 (33), probucol
  11/29 (37), warfarin probucol 1/21 (5).

29
HUMAN STUDIES

• There is increasing evidence to suggest that
  relationship between fatty marrow content and AVN
  applies to humans.
• Vande Berg showed this using MRI, that lupus
  patients with evidence of AVN on high dose
  cortocosteroids have a greater increase in the
  percentage of fatty marrow and in the index of
  marrow conversion compared to both age matched
  controls and to lupus patients on comparative
  doses of corticosteroids who did not develop AVN.
• These findings support the notion that fat
  accumulation within the closed system of the bone
  marrow produces the increased medullary pressure
  that can result in AVN.

30
The Presence of Inflammation is a Newly
Identified Risk Factor for Coronary Disease

• Increased C-reactive protein (CRP) significantly
  predicts myocardial infarction in men and women
  who did not have clinical evidence of
  atherosclerotic coronary disease.
• The risk reduction attributable to pravastatin
  was much larger among those patients with
  evidence of inflammation.
• Laboratory studies and experiments in animal
  models of atherosclerosis indicate that
  pravastatin may decrease inflammatory mediators.
• The inhibition of HMG-CoA reductase by statins
  also prevents the generation of mevalonate, the
  precursor of a complex series of isoprenoids that
  postranslationally modify certain proteins by
  isoprenylation with farnesyl or geranylgeranyl.
• The in vitro effects on endothelium (e.g.
  inhibiting adhesion molecule and iNOS expression)
  is consistent with the hypothesis that statin
  antagonism of prenylated proteins interferes with
  the activation of cellular components involved in
  the inflammatory process such as endothelial
  cells.

31
STATINS

• Inhibit HMGCo A reductase and both cholesterol
  and mevalonate synthesis (intracellular
  signalling)
• Reduce serum lipids
• Reduce risk of CAD
• Reduce risk of AVN
• Reduce risk of osteoporosis
• Modify disease activity and organ damage by
  preventing endothelial activation

32
APLLE Trial

•  Four month randomized double-blind placebo
  controlled study of atorvostatin 40 mg po qd vs
  placebo to prevent AVN in steroid treated
  patients with active SLE (rolling enrollment of
  30 patients per year for 3 years)
• Active SLE
• Steroid ? .75 mg./kg gt 4 weeks
• No current statin
• LFT lt 2x normal
• Baseline MRI 10 testable sites (bilateral femoral
  head and chondyle, tibial plateau, distal tibia
  and talus)
• Randomized to atorvostatin (lipitor) 40 mg po qd
  vs placebo
• Stratified by APS status
• Follow-up MRI 10 testable sites at 4 months and
  greater than 6 months
• Primary endpoint New AVN at any of the 10
  testable sites (i.e. 90 power to
• observe 50 reduction in new AVN at any of the 10
  testable sites per patient
• with assumption that 50 will have at least 1 new
  site of AVN)
•  
• Secondary endpoints Cholesterol, TG, HDL, LDL,
  SLEDAI, ESR, hsCRP,
• dsDNA, C3, C4, activated CEC, soluble adhesion
  molecules

33
APLLE TRIAL
34
Summary

• Enrollment n 20 15 female, 5 male 9
  African-American, 6 Hispanic, 3 Caucasian, and 2
  Asians
• 18/20 high dose steroids for renal flare
• 8/20 (40) AVN on baseline MRI secondary to
  steroid treatment for prior disease exacerbation
• MRI (4 month f/u) 13/13 no new AVN
• MRI (6 month f/u) 2/5 patients (40) demonstrate
  new AVN 1 patient with 1 new and 1 patient with
  2 new sites of AVN
• Atorvostatin well tolerated in this cohort with
  no significant elevation of transaminases or CPK

35
Conclusion

• Steroid treatment of SLE disease exacerbation is
  associated with high prevalence of AVN (i.e. 40
  (8/20) at baseline and 40 (2/5) at longer term
  MRI follow-up
• Time to onset of AVN in steroid treated patients
  uncertain (Oinuma et al AVN in patients with SLE
  develops very early after starting high dose
  steroid treatment Ann Rheum Dis
  2001601145-1148) 72 steroid SLE patients with
  MRI 1, 3, 6 and 12 months and 32/72 patients
  (44) develop ON between 1-5 months after
  starting steroids and no new AVN from 6-12 month
• Benefit of statins, atorvostatin 40 mg po qd, in
  preventing incidence of AVN and secondary
  outcomes (i.e. cholesterol, TG, LDL, HDL, ESR,
  hs-CRP, C4, C4, dsDNA, SLEDAI, Circulating
  endothelial cells) to be determined