You are on page 1of 5

31.05.

2013

Serum Amyloid A in Obstructive Sleep Apnea

C irculatio n

circ.ahajournals.org Circulation. 2003; 108: 1451- 1454 Publis hed online before print September 2, 2003, doi: 10.1161/ 01.CIR.0000089091.09527.B8

Clinical Investigation and Reports


Serum Amyloid A in Obstructive Sleep Apnea
Anna Svatik ova; Robert Wolk , MD, PhD; Abu S. Shams uzzaman, MBBS, PhD; Tomas Kara, MD; Eric J. Ols on, MD; V irend K. Somers , MD, PhD
+ Author Affiliations

Correspondence to Virend K. Somers, MD, DPhil, Mayo Clinic, Rochester, MN 55905. E-mail somers.virend@mayo.edu

Abstract
Bac k ground Patients with severe obstructive sleep apnea (OSA) may have increased risk for cardiovascular and cerebrovascular diseases. Serum amyloid A (SAA) protein has recently been linked to the development of atherosclerosis, stroke, diabetes, and dementia. We tested the hypothesis that plasma SAA levels are increased in otherwise healthy subjects with OSA. Methods and Res ults Plasma SAA levels were measured in 10 male patients with moderate to severe OSA before sleep, after 5 hours of untreated OSA, and in the morning after effective continuous positive airway pressure treatment. SAA levels were also measured in 10 closely matched control subjects at similar time points. Baseline plasma SAA levels before sleep were strikingly higher in patients with moderate to severe OSA than in controls (18.82.6 versus 7.22.6 g/mL, respectively; P=0.005) and remained unchanged in both groups throughout the night. SAA levels in 10 male patients with mild OSA were comparable with controls (P=0.46). Plasma SAA in 7 female patients with moderate to severe OSA was also markedly higher compared with matched control female subjects (24.12.4 versus 10.22.4 g/mL, respectively; P=0.0013) but was not different from male patients with moderate to severe OSA (P=0.3). There was a significant positive correlation between SAA and apnea-hypopnea index (r=0.40, P=0.03). Conc lus ions Plasma SAA levels are more than 2-fold greater in patients with moderate to severe OSA compared with subjects with mild OSA or healthy controls regardless of gender. Elevated SAA may contribute to any increased risk for cardiovascular and neuronal dysfunction in patients with OSA. Key Words :
amyloid sleep cardiovascular diseases

Received January 15, 2003; de novo received May 6, 2003; accepted July 9, 2003. Obstructive sleep apnea (OSA) is associated with increased cardiovascular 13 4 morbidity. In addition, OSA may be related to dementia. Inflammatory 4,5 mechanisms, implicated in cardiovascular and cerebral degenerative diseases, 69 may be activated in sleep disorders (including OSA) and thus contribute to the association between OSA, cardiovascular disease, and dementia. Serum amyloid A (SAA) is one of the major acute-phase proteins in humans that is upregulated by inflammatory cytokines, including interleukin-1 and interleukin10,11 6. It is synthesized predominantly by the liver and secreted as a major 10 component of the apolipoproteins in the HDL particle. Elevated SAA is 1012 associated with increased risk for coronary heart disease. SAA may accumulate in the brain of patients with Alzheimer disease, a condition that may be associated with chronic brain inflammation and may contribute to neuronal 13,14 loss and white matter damage. Any sustained elevation of SAA in patients with OSA might help explain their increased incidence of cardiovascular disease as well as dementia. We therefore tested the hypothesis that plasma SAA levels are increased in otherwise healthy subjects with OSA.

Methods
We compared subjects with moderate to severe OSA (apnea-hypopnea index [AHI] 20) with those with mild OSA (AHI 6 to 19) and those without OSA (AHI 5). We

circ.ahajournals.org/content/108/12/1451.long

1/5

31.05.2013

Serum Amyloid A in Obstructive Sleep Apnea

studied 20 male and 7 female patients with newly diagnosed OSA who were free of other diseases, had never been treated for OSA, and were taking no medications, and 10 healthy male and 7 healthy female control subjects matched for age and body mass index, in whom occult OSA was excluded by overnight polysomnography. All participants were nonsmokers. The presence and severity of sleep apnea were determined by standard overnight polysomnography. AHI was calculated as the total number of apneas and hypopneas per hour of sleep. Sleep studies followed a split-night protocol according to the standard of care in our institution. The first half of the study was for the diagnosis of OSA. A therapeutic continuous positive airway pressure (CPAP) trial followed in the second half of the night. SAA levels in 10 male patients with moderate to severe OSA were measured at 9:00 PM (before sleep), at 2:00 AM (after 5 hours of untreated OSA, before CPAP therapy started), and at 6:00 AM (after waking in the morning, after 4-hour CPAP treatment). Measurements were obtained at similar times in 10 control subjects. Plasma SAA was also measured at baseline, before sleep, in 10 male patients with mild OSA, 7 female patients with moderate to severe OSA, and 7 matched healthy female control subjects. The study was approved by the Human Subjects Review Committee. Plasma SAA concentrations were measured using ELISA (SAA kit, BioSource International, Inc) using a monoclonal antibody specific for SAA. The minimum detectable concentration of SAA was 1.3 g/mL. Intraassay coefficients of variation were 3.1%, 6.0%, and 12% at 15.7, 71.5, and 192 g/mL, respectively. Interassay coefficients of variation were 12%, 13%, and 11% at 7.2, 36.1, and 70.5 g/mL, respectively. Spike recovery of SAA yielded an average recovery of 94%. Results are reported as meanSEM. Continuous variables were compared between groups using one-way ANOVA. Differences between various time points (9:00 PM, 2:00 AM, and 6:00 AM) within each group were assessed using ANOVA for repeated measures, followed by the Newman-Keuls post-hoc tests. Statistical significance was defined as P<0.05.

Results
The OSA and control groups were very similar with regard to demographics, hemodynamics, and metabolic characteristics (Table). OSA subjects suffered more severe hypoxic burden than control subjects (Table). Treatment with CPAP in the OSA group reduced AHI from 464 to 104 events per hour (P=0.002) and increased O2 nadir from 811% to 921% (P=0.0002). Time spent at O2 saturation below 90% decreased from 123 to 03 minutes (P=0.02).

Bas eline Charac teris tic s of Patients With OSA and Control Subjec ts

View this table: In this window In a new window

Baseline plasma SAA levels before sleep were strikingly higher in male patients with moderate to severe OSA than in control subjects (18.82.6 versus 7.22.6 g/mL, respectively; P=0.005). These SAA levels remained markedly higher but stable throughout the night in the OSA group (between group comparison, F=7.8; P=0.01) and were not affected by several hours of untreated OSA or by acute CPAP treatment (Figure 1). SAA levels also remained stable through the night in the control subjects. Baseline SAA level in 10 male patients with mild OSA was comparable with the control group (8.72.6 versus 7.22.6 g/mL, respectively; P=0.46) (Figure 2). Plasma SAA in 7 female patients with moderate to severe OSA was markedly higher compared with 7 closely matched female control subjects (24.12.4 versus 10.22.4 g/mL, respectively; P=0.0013) (Figure 2) but not different from male patients with moderate to severe OSA (P=0.3). There was a significant positive correlation between SAA and AHI (r=0.40, P=0.03).

View larger version: In this page In a new window Download as PowerPoint Slide

Figure 1. Plasma SAA in 10 male patients with OSA before sleep, after 5 hours of untreated OSA, and after 4 hours of CPAP compared with 10 control subjects in whom blood samples were taken at similar time points. CPAP was applied between 2:00 AM and 6:00 AM. Data are meanSEM. P=0.01 for between-group comparison.

circ.ahajournals.org/content/108/12/1451.long

2/5

31.05.2013

Serum Amyloid A in Obstructive Sleep Apnea

View larger version: In this page In a new window Download as PowerPoint Slide

Figure 2. Baseline SAA levels in 10 male patients with moderate to severe OSA and 10 male patients with mild OSA compared with 10 matched healthy male control subjects. Plasma SAA in 7 female patients with moderate to severe OSA compared with 7 healthy female control subjects. Data are mean SEM.

Discussion
The novel findings of our present study include, first, that patients with moderate to severe OSA have elevated SAA levels, 2.5-fold greater than measurements in healthy subjects or patients with mild OSA. The mean level of SAA in patients with OSA without other disease is even higher than the 75th percentile of SAA measurements obtained in patients with unstable angina or nonQ-wave myocardial infarction enrolled in the Thrombolysis In Myocardial Infarction 11A 12 study. Second, untreated OSA, CPAP treatment, or normal sleep did not acutely affect SAA levels. Effects of chronic treatment with CPAP on SAA are unknown and remain to be established. The mechanisms whereby OSA affects SAA levels are unknown. One possibility is the effect of hypoxia/reoxygenation related to sleep apnea. Hypoxia stimulates the genes of acute-phase proteins, as well as cytokines known to induce these 15 proteins. Hypoxemia at high altitude may increase plasma levels of several 16 inflammatory mediators. Sleep deprivation or fragmentation may also influence 68 systemic inflammation, including elevation of circulating cytokine levels. Patients with OSA in our study were characterized not only by severe apnea (Table) but also by a much higher arousal index, suggesting that both hypoxemia and sleep fragmentation may be implicated in elevated SAA levels. Nevertheless, these perturbations do not seem to have any acute effects on SAA through the night but may be linked to chronic SAA elevations evident even before sleep. The association between OSA and elevated SAA may have important pathophysiological implications. Compelling epidemiological evidence implicates 1012 13,14 increased SAA in cardiovascular and neurological diseases, suggesting that SAA may be an important link between OSA and conditions such as atherosclerosis, diabetes, and dementia. Atherosclerosis, Inflam m ation, and Metabolic Dy sfunction Low-grade inflammation contributes importantly to the initiation and the 5 progression of atherogenic processes. Elevated SAA constitutes part of a general inflammatory response and may also exert direct proatherogenic effects. For example, the association of SAA with HDL might alter the metabolism of this lipoprotein particle, thereby compromising its protective effect against 10 atherosclerosis. Recent observations suggest that SAA can predict early mortality in patients with stable and unstable angina, myocardial infarction, and coronary artery disease as well as in patients with other coronary and peripheral 11,12 vascular disease. Sleep loss and hypoxemia may also be causally associated with metabolic 17,18 derangements. Patients with sleep-disordered breathing have insulin 2 resistance, with higher fasting glucose and insulin levels. SAA tissue expression and a hepatic receptor for SAA (Tanis) are dysregulated in association with glucose 19,20 intolerance. Whether elevated SAA is related to glucose intolerance in OSA remains unclear. Dem entia An etiological link between OSA and dementia, although unproven, is suggested 4 by the observation that CPAP treatment can reduce the level of dementia. Amyloid 13 deposition is thought to play an important role in many forms of dementia. Elevated levels of SAA may be related to tissue amyloid deposition. Proteolysis of SAA to amyloid A may lead to formation of amyloid fibrils, which can be seen on histological examinations of brain tissues obtained from subjects with 14 dementia. In Alzheimer disease, amyloid (the main component of amyloid 13 plaque) can be found in biological fluids as a soluble protein. Elevated SAA in patients with OSA may conceivably be related to some types of dementia seen in OSA.

circ.ahajournals.org/content/108/12/1451.long

3/5

31.05.2013
Strengths of the Study

Serum Amyloid A in Obstructive Sleep Apnea

First, we included only normotensive, newly diagnosed, and untreated OSA patients who had no coexisting diseases apart from OSA. Second, patients and controls were not taking any medications. Third, the control subjects were matched for age and body mass index, thus ruling out any potential confounding influence of age and especially obesity on our data. Fourth, complete overnight polysomnography excluded any possibility of occult sleep apnea in our obese control subjects. Conclusion SAA levels are chronically elevated in patients with OSA. SAA has been associated with cardiovascular and neuronal diseases and may be an important potential mechanism to explain any increased risk for cardiovascular and neuronal dysfunction in patients with OSA.

Acknowledgments
This study was supported by NIH grants HL-65176, HL-61560, HL-70602, and M01-RR00585.

References
1. Peker Y, Kraiczi H, Hedner J, et al. An independent association between obstructive sleep apnoea and coronary artery disease. Eur Respir J. 1999; 14: 179184. A bs tr ac t/ FREE Fu ll Tex t 2. Ip MS, Lam B, Ng NM, et al. Obstructive sleep apnea is independently associated with insulin resistance. Am J Respir Crit Care Med. 2002; 65: 670 676. 3. Bassetti C, Aldrich MS, Quint D. Sleep-disordered breathing in patients with acute supra- and infratentorial strokes: a prospective study of 39 patients. Stroke. 1997; 28: 17651772. A bs tr ac t/ FREE Fu ll Tex t 4. Bliwise DL. Sleep apnea, APOE4 and Alzheimers disease 20 years and counting? J Psychosom Res . 2002; 53: 539546. C r o s s Ref Medlin e 5. Ridker PM, Cushman M, Stampfer MJ, et al. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med. 1997; 336: 973979. C r o s s Ref Medlin e 6. Mullington JM, Hinze-Selch D, Pollmacher T. Mediators of inflammation and their interaction with sleep: relevance for chronic fatigue syndrome and related conditions. Ann N Y Acad Sci . 2001; 933: 201210. Medlin e 7. Shearer WT, Reuben JM, Mullington JM, et al. Soluble TNF-alpha receptor 1 and IL-6 plasma levels in humans subjected to the sleep deprivation model of spaceflight. J Allergy Clin Immunol . 2001; 107: 165170. C r o s s Ref
Medlin e

8. Rogers NL, Szuba MP, Staab JP, et al. Neuroimmunologic aspects of sleep and sleep loss. Semin Clin Neuropsychiatry . 2001; 6: 295307. C r o s s Ref
Medlin e

9. Shamsuzzaman AS, Winnicki M, Lanfranchi P, et al. Elevated C-reactive protein in patients with obstructive sleep apnea. Circulation. 2002; 105: 24622464. A bs tr ac t/ FREE Fu ll Tex t 10. Kisilevsky R, Tam SP. Acute phase serum amyloid A, cholesterol metabolism, and cardiovascular disease. Pediatr Pathol Mol Med. 2002; 21: 291305.
C r o s s Ref Medlin e

11. Delanghe JR, Langlois MR, De Bacquer D, et al. Discriminative value of serum amyloid A and other acute-phase proteins for coronary heart disease. Atherosclerosis . 2002; 160: 471476. C r o s s Ref Medlin e 12. Morrow DA, Rifai N, Antman EM, et al. Serum amyloid A predicts early mortality in acute coronary syndromes: a TIMI 11A substudy. J Am Coll Cardiol . 2000; 35: 358362. A bs tr ac t/ FREE Fu ll Tex t 13. Ghiso J, Wisniewski T, Frangione B. Unifying features of systemic and cerebral amyloidosis. Mol Neurobiol . 1994; 8: 4964. Medlin e 14. Chung TF, Sipe JD, McKee A, et al. Serum amyloid A in Alzheimers disease brain is predominantly localized to myelin sheaths and axonal membrane. Amyloid. 2000; 7: 105110. Medlin e 15. Wenger RH, Rolfs A, Marti HH, et al. Hypoxia, a novel inducer of acute phase gene expression in a human hepatoma cell line. J Biol Chem. 1995; 270: 2786527870. A bs tr ac t/ FREE Fu ll Tex t 16. Hartmann G, Tschop M, Fischer R, et al. High altitude increases circulating interleukin-6, interleukin-1 receptor antagonist and C-reactive protein. Cytokine. 2000; 12: 246252. C r o s s Ref Medlin e 17. Tasali E, Van Cauter E. Sleep-disordered breathing and the current epidemic of obesity: consequence or contributing factor? Am J Respir Crit Care Med. 2002; 165: 562563. FREE Fu ll Tex t 18. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on metabolic and endocrine function. Lancet . 1999; 354: 14351439. C r o s s Ref Medlin e

circ.ahajournals.org/content/108/12/1451.long

4/5

31.05.2013

Serum Amyloid A in Obstructive Sleep Apnea

19. Walder K, Kantham L, McMillan JS, et al. Tanis: a link between type 2 diabetes and inflammation? Diabetes . 2002; 51: 18591866. A bs tr ac t/ FREE Fu ll Tex t 20. Lin Y, Rajala MW, Berger JP, et al. Hyperglycemia-induced production of acute phase reactants in adipose tissue. J Biol Chem. 2001; 276: 4207742083.
A bs tr ac t/ FREE Fu ll Tex t

Articles citing this article


Adaptive s ervoventilation in dias tolic heart failure and Cheyne- Stok es res piration Eur Respir J. 2010;36:385-392, Abs trac t Full Text PDF Sleep Apnea and Cardiovas c ular Dis eas e: An Americ an H eart As s oc iation/Americ an College of Cardiology Foundation Sc ientific Statement From the Americ an H eart As s oc iation Counc il for H igh Blood Pres s ure Res earc h Profes s ional Educ ation Committee, Counc il on Clinic al Cardiology, Strok e Counc il, and Counc il on Cardiovas c ular N urs ing In Collaboration With the N ational H eart, Lung, and Blood Ins titute N ational Center on Sleep Dis orders Res earc h (N ational Ins titutes of H ealth) Circulation. 2008;118:1080-1111, Full Text PDF Metabolic dis turbanc es in patients with obs truc tive s leep apnoea s yndrome Eur Respir Rev. 2007;16:196-202, Abs trac t Full Text PDF Sleep Apnoea & H ypertens ion: Phys iologic al bas es for a c aus al relation: Sleep and the metabolic s yndrome Exp Physiol. 2007;92:67-78, Abs trac t Full Text PDF The vas c ular mic romilieu in obs truc tive s leep apnoea Eur Respir J. 2005;25:780-782, Full Text PDF Circ ulating free nitrotyros ine in obs truc tive s leep apnea Am. J. Physiol. Regul. Integr. Comp. Physiol.. 2004;287:R284-R287, Abs trac t Full Text PDF Pro: Sleep Apnea Caus es Cardiovas c ular Dis eas e Am. J. Respir. Crit. Care Med.. 2004;169:147-148, Full Text PDF

circ.ahajournals.org/content/108/12/1451.long

5/5