- Email: [email protected]
Acute phase reactants in patients with abdominal aortic aneurysm
Atherosclerosis 163 (2002) 297 /302 www.elsevier.com/locate/atherosclerosis
Acute phase reactants in patients with abdominal aortic aneurysm H. Domanovits a,*, M. Schillinger b, M. Mu¨llner a, T. Ho¨lzenbein c, K. Janata a, K. Bayegan a, A.N. Laggner a a
Department of Emergency Medicine, Vienna General Hospital, University of Vienna, Waehringer Guertel 18-20/6D, 1090 Vienna, Austria b Department of Angiology, Vienna General Hospital, University of Vienna, Waehringer Guertel 18-20/6D, 1090 Vienna, Austria c Department of Vascular Surgery, Vienna General Hospital, University of Vienna, Waehringer Guertel 18-20/6D, 1090 Vienna, Austria Received 18 April 2001; received in revised form 7 December 2001; accepted 13 December 2001
Abstract Background: There is increasing evidence that an inflammatory process is present in abdominal aortic aneurysms (AAAs) to varying degrees. The aim of this study was to compare acute phase reactants in patients with asymptomatic AAA, symptomatic AAA without rupture and ruptured AAA. Method: Two hundred and twenty-five consecutive patients treated because of AAA were included in this case-control study. Polynomial logistic regression analysis was applied to compare admission C-reactive protein (CRP) and white blood count (WBC) measured in 111 asymptomatic outpatients, 52 symptomatic patients without rupture and 62 patients with rupture of the aneurysm. We adjusted for the potentially confounding effect of age, sex, haemoglobin levels and aneurysm diameter. Results: Patients with symptomatic AAA and patients with ruptured AAA had significantly elevated CRP (p / 0.002) and WBC (p B/0.0001) levels compared to asymptomatic patients. There was no statistically significant difference in CRP and WBC between patients with symptomatic AAA and ruptured AAA. Median CRP values of asymptomatic, symptomatic and ruptured AAA were B/0.5 (interquartile range (IQR) B/0.5 /0.85), 1.1(IQR B/0.5 /4.0) and 2.4 mg/dl (IQR 0.65 /8.6), respectively, and median WBC values were 6.5 (IQR 5.5 /8.0), 8.7 (IQR 6.8 /11.2) and 13.2 (IQR 10.5 /17.0), respectively. Conclusion: A significant elevation of CRP and WBC could be found in patients who presented with symptoms or rupture of an AAA. These indicators of inflammation were not observed in asymptomatic patients with AAA. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Abdominal aortic aneurysm; Inflammation; CRP; Rupture
1. Introduction There is increasing evidence that inflammatory process is present in abdominal aortic aneurysms (AAAs) to varying degrees [1 /5]. Recent reports suggest that the inflammatory aortic aneurysm is not a distinct clinical entity and arises from the same pathogenic stimulus responsible to non-inflammatory AAAs [1]. A gradual passage in terms of inflammatory response from ordinary atherosclerotic to inflammatory AAA was described [4].
* Corresponding author. Tel.: 43-1-40-400-1964; fax: 43-1-40400-3953. E-mail address: [email protected] (H. Domanovits).
The degradation of the elastic media is a hallmark in the development of AAAs, a process which may be triggered by inflammatory response [6]. Acute phase reactants are expected to play a role in the progression of aortic disease [7]. However, the association of laboratory parameters of inflammation like C-reactive protein (CRP), white blood count (WBC) and fibrinogen with the clinical presentation and acuity of aortic aneurysm have not been investigated up to now. An increase of acute phase reactants in patients with ruptured AAA can be expected. However, signs of an underlying inflammatory process in symptomatic patients without rupture have not been evaluated. The aim of the study was to determine the association of acute phase reactants CRP, WBC and fibrinogen in patients with asymptomatic, symptomatic and ruptured AAA.
0021-9150/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 2 1 - 9 1 5 0 ( 0 2 ) 0 0 0 0 6 - 0
298
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
2. Methods 2.1. Patients Patients with symptomatic AAA without signs of rupture and patients with ruptured AAA were compared to patients with asymptomatic AAA. We included all consecutive patients with a symptomatic AAA admitted to the emergency department (ED) of a 2200 bed university hospital between 1 January, 1992 and 31 November, 1998. The comparison cohort comprised a group of consecutive patients with asymptomatic AAA treated at the outpatient unit of the Department of Vascular Surgery within the same time interval. These patients were scheduled for elective endovascular or surgical aneurysm repair. Patients with inflammatory aneurysm [8,9] based on computed tomography findings were not included. Concomitant inflammatory diseases which obviously cause elevation of acute phase reactants were excluded by clinical examination, chest radiography and urinalysis. Seven patients (6%) were not included in the final analysis because of coexistent inflammatory disease such as pneumonia (n /2), urinary tract infection (n /4) and cholecystitis (n /1). 2.2. Definitions An AAA was defined as a permanent dilatation of the aorta, with a diameter of at least 50% greater than the proximal aortic neck [10]. Rupture was defined by direct visualisation of discontinuity of the aortic wall by computed tomography, sonography, intraoperative findings or at autopsy. Symptomatic aneurysm was defined if a patient presented with pain, referred as lower back pain, abdominal or flank pain, which was supposed to be originating from the aneurysm. 2.3. Laboratory parameters Serum CRP level, WBC and plasma fibrinogen are routinely measured at the time of admission to the emergency department and during routine visits to the outpatient ward of the Department of Vascular Surgery. For measurement of serum CRP levels Tina-quant† (Roche) was used, WBC was measured by NE 9000† (Sysmex) and fibrinogen was measured by Fibrinogen Clauss† (Stago/Roche). 2.4. Follow up Patients who were treated conservatively were followed for 6 months for the occurrence of ruptureassociated death. These follow-up data were derived from the Austrian Central Statistical Office (Day and Cause of Death, ICD 9).
The study was performed according to the Declaration of Helsinki. 2.5. Statistical analysis Continuous data are presented as the median and the interquartile range (IQR, range from the 25th to 75th percentile). Percentages were determined for dichotomous variables. For univariate comparison of age, sex haemoglobin and aneurysm diameter between the groups the x2-test and the Kruskal Wallis test was used. Spearman’s rank correlation coefficient was calculated to assess the strength of association between continuous variables. Polynomial logistic regression models were applied to assess the independent association of CRP, WBC and fibrinogen between the three groups (symptomatic AAA vs. ruptured AAA vs. asymptomatic AAA) and to adjust for patients, age, sex and haemoglobin levels. A P -value 5/0.05 was accepted as statistically significant. All calculations were performed with Excel for Windows, and SPSS (Version 10.0) for Windows.
3. Results Within the observation period 114 eligible patients presented at the ED because of symptomatic AAA. In 62 of these patients rupture of the AAA was diagnosed. The median aneurysm diameter in patients with ruptured aneurysm was 80 mm (IQR 70 /90), the median aneurysm diameter in patients with symptomatic aneurysms without rupture was 60 mm (IQR 50/70). The comparison group comprised 111 patients from the outpatient ward of the Department of Vascular Surgery with asymptomatic AAA with a median aneurysm diameter of 54 mm (IQR 50 /63) (P B/0.0001). Median age of all patients was 72 years (IQR 66 /76) and 187 patients (83%) were male. Baseline data and laboratory findings of the three groups are presented in Table 1. Age was similar in the three groups (P /0.7), the sex distribution differed significantly (P /0.009). Haemoglobin was also significantly different, patients with ruptured AAA had the lowest levels (P B/0.0001). Differences of baseline CRP and WBC levels were found between the three groups using univariate logistic regression (P B/0.0001) (Table 2). CRP and WBC were both taken as an indicator for acute phase reaction, but were not analysed in the same multivariate model for reasons of collinearity. Two separate multivariate polynomial logistic regression models were applied to assess the independent association of CRP and WBC with the group classification of AAA and to adjust for patients age, sex, haemoglobin levels and aneurysm diameter. Compared to asymptomatic patients, CRP and WBC were independently
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
299
Table 1 Baseline data of 225 patients with AAA
Median age in years (IQR) Male sex (%) Median haemoglobin (g/dl) (IQR) Median CRP (mg/dl) (IQR) WBC per 1000 (IQR) Fibrinogen (mg/dl) (IQR)
Asymptomatic AAA (n 111)
Symptomatic AAA (n 52)
Ruptured AAA (n 62)
73 (67 /76) 100 (90%)* 14.3 (13.2 /15.2)* B 0.5 ( B 0.5 /0.85)* 6.5 (5.5 /8.0)* 327 (273 /392)
72 (67 /77) 37 (71%)* 12.2 (11.1 /14.5)* 1.1 ( B 0.5 /4.0)* 8.7 (6.8 /11.2)* 388 (308 /466)
72 (65 /77) 50 (81%)* 10.3 (9.2 /12.6)* 2.4 (0.65 /8.6)* 13.2 (10.5 /17.0)* 309 (216 /433)*
AAA, abdominal aortic aneurysm; IQR, interquartile range; CRP, C-reactive protein; WBC, white blood count. * P B 0.05.
elevated in patients with symptomatic disease and in patients with ruptured AAA after adjusting for age and sex. When additionally adjusting for haemoglobin and aneurysm diameter, minimal changes of odds ratios were observed. The difference of CRP and WBC between the three groups, however, remained highly significant (Tables 3 /5). When additionally adjusting for aneurysm diameter the odds ratios and the 95% confidence interval (CI) remained largely unchanged. Therefore, it seems unlikely that CRP merely reflects the aneurysm size. Furthermore, in symptomatic patients, there was no correlation between the aneurysm diameter and the CRP level (Spearman’s rank correlation r / 0.06, P /0.8). To answer the question of the relative contribution of low haemoglobin vs. high CRP in predicting those with symptoms, we tested for interaction between these variables in the polynomial model: when CRP and haemoglobin were entered into the model with and without a multiplicative interaction term, no significant alteration of the model fit was observed, thus we believe that CRP does not vary over levels of haemoglobin. Table 2 Association of serum CRP levels, WBC and fibrinogen with the clinical presentation of AAA in univariate polynomial logistic regression models: asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA OR
95% CI
P -value
CRP Asymptomatic AAA (n 111) Symptomatic AAA (n 52) Ruptured AAA (n 62)
1.0 1.3 1.4
/ 1.2 /1.6 1.2 /1.6
/ B 0.0001 B 0.0001
WBC Asymptomatic AAA (n 111) Symptomatic AAA (n 52) Ruptured AAA (n 62)
1.0 1.4 1.8
/ 1.3 /1.7 1.5 /2.1
/ B 0.0001 B 0.0001
Fibrinogen Asymptomatic AAA (n 111) Symptomatic AAA (n 52) Ruptured AAA (n 62)
1.0 1.0 1.0
/ 1.0 /1.0 1.0 /1.0
/ 0.05 0.6
AAA, abdominal aortic aneurysm; IQR, interquartile range; CRP, Creactive protein; WBC, white blood count; OR, odds ratio; CI, confidence interval; /, denotes reference group.
We applied a similar model using the variable fibrinogen as an indicator for the acute phase reaction, which showed no significant differences between the three groups (data not shown) (data presented in the appendix for information of editors and reviewers). Fifty-two patients with initially conservative treatment were followed for up to 6 months to investigate the Table 3 Independent association of serum CRP levels with asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA in a multivariate polynomial logistic regression model OR
95% CI
P -value
Model controlled for age (years ) and sex Asymptomatic AAA (n 111) 1.0 Symptomatic AAA (n 52) 1.3 Ruptured AAA (n 62) 1.4
/ 1.2 /1.5 1.2 /1.6
/ B 0.0001 B 0.0001
Model controlled for age (years ), sex , haemoglobin (g /dl ) and aneurysm diameter (cm ) Asymptomatic AAA (n 111) 1.0 / / Symptomatic AAA (n 52) 1.3 1.1 /1.5 0.008 0.007 Ruptured AAA (n 62) 1.3 1.1 /1.5 AAA, abdominal aortic aneurysm; IQR, interquartile range; OR, odds ratio; CI, confidence interval; /, denotes reference group.
Table 4 Independent association of WBC with asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA in a multivariate polynomial logistic regression model OR
95% CI
P -value
Model controlled for age (years ) and sex Asymptomatic AAA (n 111) 1.0 Symptomatic AAA (n 52) 1.5 Ruptured AAA (n 62) 1.8
/ 1.3 /1.7 1.5 /2.1
/ B 0.0001 B 0.0001
Model controlled for age (years ), sex , haemoglobin (g /dl ) and aneurysm diameter (cm ) Asymptomatic AAA (n 111) 1.0 / / Symptomatic AAA (n 52) 1.8 1.4 /2.2 B 0.0001 B 0.0001 Ruptured AAA (n 62) 2.0 1.5 /2.5 AAA, abdominal aortic aneurysin; IQR, interquartile range; OR, odds ratio; CI, confidence interval; /, denotes reference group.
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
300
Table 5 Independent association of fibrinogen with asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA in a multivariate polynomial logistic regression model OR
95% CI
P -value
Model controlled for age (years ) and sex Asymptomatic AAA (n 111) 1.0 Symptomatic AAA (n 52) 1.0 Ruptured AAA (n 62) 1.0
/ 1.0 /1.0 1.0 /1.0
/ 0.04 0.9
Model controlled for age (years ), sex , haemoglobin (g /dl ) aneurysm diameter (cm ) Asymptomatic AAA (n 111) 1.0 / Symptomatic AAA (n 52) 1.0 1.0 /1.0 Ruptured AAA (n 62) 1.0 1.0 /1.0
and / 0.1 0.4
AAA, abdominal aortic aneurysm; IQR, interquartile range; OR, odds ratio; CI, confidence interval, /, denotes reference group.
association of baseline CRP and WBC levels and the occurrence of death due to aneurysm rupture. Eleven patients died due to aneurysm rupture during this time interval. These patients had significantly higher baseline CRP levels (median 3.9 mg/dl, IQR 1.6 /13.3) compared to patients without death due to rupture (median 0.9 mg/dl, IQR B/0.5 /1.8) (P /0.006). Furthermore, patients with death due to rupture had significantly higher baseline WBC (median 10.4/1000, IQR 8.8 /14.1) compared to survivors without rupture (median 7.6/1000, IQR 6.2 /10.5) (P /0.004).
4. Discussion We found higher serum CRP and WBC levels in patients with symptomatic and ruptured AAA compared to asymptomatic patients. Not surprisingly elevated laboratory parameters of inflammation were present in patients with ruptured aneurysm. Remarkably, there was also a significant elevation in symptomatic patients without rupture. Others have shown aneurysmal pain to be associated with expansion or wall distension with impending rupture or dissection [11 /13]. We found aneurysmal pain associated with an elevation of CRP and WBC. However, we could not document expansion or wall distension due to the cross sectional design of our study. As aneurysmal pain was associated with elevated CRP and WBC, our findings suggest, that an inflammatory process is present in the aortic wall which promotes expansion and rupture. This inflammatory process may be quantified by measurement of serum CRP and WBC. 4.1. Inflammatory activation Inflammatory activation is an important pathophysiologic process in several atherosclerotic entities like
coronary artery disease, peripheral artery disease and aortic aneurysm [1 /4,9,14/17]. Inflammatory cytokines were shown to be elevated in patients with AAA [18], which lead to the degradation of extracellular matrix components [19,20], a basic feature in the pathogenesis of AAA. The cause of the inflammatory process in the aneurysmal wall is unknown; it may be an autoimmune reaction as well as an infectious agent like chlamydia pneumoniae [21]. The elevation of CRP, WBC and fibrinogen may be an indicator of acute phase response due to vascular tissue damage and inflammation in atherosclerotic disease [22 /24]. However, so far it cannot be distinguished to what degree the elevated serum markers of inflammation in patients with acute aortic aneurysm are due to local vascular inflammatory process or due to acute phase reaction accompanying the systemic inflammatory response syndrome.
4.2. Diagnostic value The diagnostic value of CRP and WBC in patients with AAA has not been evaluated. Concomitant disorders with overlapping signs and symptoms occur frequently in the elderly population with AAA [25,26]. Abdominal, flank or back pain are the most common symptoms in patients with an expanding or ruptured AAA [11,12]. However, patients with expanding AAA may also present with a paucity of symptoms and even with ruptured AAA patients may fail to manifest signs of hemodynamic instability upon rupture [27,28]. Signs and clinical symptoms in patients with AAA are sometimes not a reliable indicator for the acuity of the disease. Acute phase reaction as a measurable parameter might be useful in differentiation of patients with and without acute disease.
4.3. Prognostic value The prognostic value of elevated CRP and WBC in patients with symptomatic AAA may be discussed: during up to 6 months follow-up period, 11 out of 52 patients with symptomatic AAA without rupture died due to rupture of the aneurysm. Significantly higher levels of CRP and WBC were found at presentation in the symptomatic non-survivors compared to symptomatic survivors. However, larger patient samples who are primarily scheduled for conservative treatment will be necessary to confirm these findings, since the majority of patients included in the present study was operated upon the aneurysm during the follow-up period and therefore was not followed for aneurysm rupture or death.
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
4.4. Therapeutic implication Chronic inflammation indicated by laboratory parameters may play a role in the progression of aortic disease [7]. Experimental approaches with anti-inflammatory therapy successfully inhibited the progress of disease [29 /31]. Particularly, the control of inflammatory response via suppression of proteolytic enzymes showed promising results in animal models [30,31]. Comparable data in humans are not available. Accordingly, the success of different conservative therapeutic strategies might be measured by monitoring the course of inflammation markers. 4.5. Limitations For measurement of serum CRP levels a standard CRP assay was used. High sensitivity CRP assays are required for atherosclerotic risk prediction in apparently healthy individuals [32]. Patients in this study with symptomatic disease had median baseline CRP levels of 1.1 and 2.4 mg/dl, which is significantly above the detection level (0.3 mg/dl) of the used assay. Therefore, the standard assay seems to be appropriate for this setting. 4.6. Conclusion An increase of CRP and WBC could be found in patients with symptomatic AAA. Remarkably, even in symptomatic patients without ruptures significant acute phase reaction was observed.
Acknowledgements The authors thank Professor Dr E. Schuster from the Institute for Medical Computer Sciences for statistical advice.
References [1] Rasmussen TE, Hallett JW. New insights into inflammatory aortic aneurysms. Eur J Vasc Endovasc Surg 1997;14:329 /32. [2] Rose AG, Dent DM. Inflammatory variant of abdominal aortic atherosclerotic aneurysm. Arch Pathol Lab Med 1981;105:409 / 13. [3] Pennel RC, Hollier LH, Lie JT. Inflammatory abdominal aortic aneurysms: a thirty year review. J Vasc Surg 1985;2:859 /69. [4] Sterpetti AV, Hunter WJ, Feldhaus RJ. Inflammatory aneurysm of the abdominal aorta: incidence, pathologic, and etiologic considerations. J Vasc Surg 1989;9:643 /50. [5] Rehm JP, Grange JJ, Baxter BT. The formation of aneurysms. Semin Vasc Surg 1998;11:193 /202.
301
[6] Davis V, Persidskaia R, Baca Regen L, Itoh Y, Nagase H, Persidsky Y, et al. Matrix metalloproteinase-2 production and its binding to the matrix are increased in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol 1998;18:1625 /33. [7] Rohde LE, Arroyo LH, Rifai N, Creager MA, Libby P, Ridker PM, et al. Plasma concentrations of interleukin-6 and abdominal aortic diameter among subjects without aortic dilatation. Arterioscler Thromb Vasc Biol 1999;19:1695 /9. [8] Walker DI, Bloor K, Williams G, Gillie I. Inflammatory aneurysms of the abdominal aorta. Br J Surg 1972;59:609 /14. [9] Rasmussen TE, Hallett JW, Jr.. Inflammatory abdominal aortic aneurysms: a review with new perspectives in etiology. Ann Surg 1997;225:1 /10. [10] Johnston KW, Rutherford RB, Tilson MD, Shah DM, Hollier L, Stanley JC. Suggested standard for reporting on arterial aneurysms. J Vasc Surg 1991;13:452 /8. [11] Vohra RD, Groome J. Long-term survival in patients undergoing resection of abdominal aortic aneurysm. Ann Vasc Surg 1990;4:460 /5. [12] Darke SG, Eadie DDG. Abdominal aortic aneurysmectomy: a review of 60 consecutive cases contrasting elective and emergency surgery. J Cardiovasc Surg 1973;14:484 /91. [13] Sullivan CA, Rohrer MJ, Cutler BS. Clinical management of the symptomatic but unruptured abdominal aortic aneurysm. J Vasc Surg 1990;11:799 /803. [14] Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Plasma concentration of C-reactive protein and risk of developing peripheral vascular disease. Circulation 1998;97:425 / 8. [15] Heinrich J, Schulte H, Schoenfeld R, Koehler E, Assmann G. Association of variables of coagulation, fibrinolysis and acutephase with atherosclerosis in coronary and peripheral arteries and those arteries supplying the brain. Thromb Haemostasis 1995;73:374 /9. [16] Liuzzo G, Biasucci L, Gallimore JR, Grillo RL, Rebuzzi AG, Pepys MB, et al. The prognostic value of C-reactive protein and serum amyloid A protein in severe unstable angina. N Engl J Med 1994;331:417 /24. [17] Mendall MA, Patel P, Ballam L, Strachan D, Northfield TC. C reactive protein and its relation to cardiovascular risk factors: a population based cross sectional study. Br Med J 1996;312:1061 / 5. [18] Juvonen J, Surcel HM, Satta J, Teppo AM, Bloigu A, Syrjala H, et al. Elevated circulating levels of inflammatory cytokines in patients with abdominal aortic aneurysm. Atheroscler Thromb Vasc Biol 1997;17:2843 /7. [19] Satta J, Laurila A, Paakko P, Haukipuro K, Sormunen R, Parkkila S, et al. Chronic inflammation and elastin degradation in abdominal aortic aneurysm disease: an immunohistochemical and electron microscopic study. Eur J Vasc Endovasc Surg 1998;15:313 /9. [20] Grange JJ, Davis V, Baxter BT. Pathogenesis of abdominal aortic aneurysm: an update and look toward the future. Cardiovasc Surg 1997;5:256 /65. [21] Lindholt JS, Fasting H, Henneberg EW, Ostergaard L. A review of chlamydia pneumoniae and atherosclerosis. Eur J Vasc Endovasc Surg 1999;17:283 /9. [22] Danesh J, Collins R, Appleby P, Peto R. Fibrinogen, C-reactive protein, albumin or white cell count: meta-analyses of prospective studies of coronary heart disease. J Am Med Assoc 1998;279:1477 /82. [23] Ridker PM. C-reactive protein and risks of future myocardial infarction and thrombotic stroke. Eur Heart J 1998;19:1 /3. [24] Maseri A, Biasucci LM, Liuzzo G. Inflammation in ischemic heart disease. Br Med J 1996;312:1049 /50. [25] Rothrock SG, Greenfield RH, Falk JL. Acute abdominal emergencies in the elderly: clues to identifying serious illness.
302
[26]
[27]
[28]
[29]
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302 Part I */clinical assessment and diagnostic studies. Emerg Med Rep 1992;13:177 /84. Rothrock SG, Greenfield RH, Falk JL. Acute abdominal emergencies in the elderly: clinical evaluation and management. Part II */diagnosis and management of common conditions. Emerg Med Rep 1992;13:185 /92. Zimmer T. Absence of back pain and tachycardia in the emergency presentation of abdominal aortic aneurysm. Am J Emerg Med 1988;6:316 /21. Bower TC, Cherry KJ, Pairolero PC. Unusual manifestations of abdominal aortic aneurysms. Surg Clin North Am 1989;69:745 / 54. Miralles M, Wester W, Sicard GA, Thompson R, Reilly JM. Indomethacin inhibits expansion of experimental aortic aneurysm
via inhibition of the cox2 isoform of cyclooxygenase. J Vasc Surg 1999;29:884 /92. [30] Curci JA, Petrinec D, Liao S, Golub LM, Thompson RW. Pharmacologic suppression of experimental aortic aneurysms: a comparison of doxycycline and four chemically modified tetracyclines. J Vasc Surg 1998;28:1082 /93. [31] Bigatel DA, Elmore JR, Carey DJ, Cizmeci Smith G, Franklin DP, Youkey JR. The matrix metalloproteinase inhibitor BB-94 limits expansion of experimental abdominal aortic aneurysms. J Vasc Surg 1999;29:130 /8. [32] Roberts WL, Sedrick R, Moulton L, Spencer A, Rifai N. Evaluation of four automated high-sensitivity C-reactive protein methods: implications for clinical and epidemiological applications. Clin Chem 2000;46:461 /8.
Acute phase reactants in patients with abdominal aortic aneurysm H. Domanovits a,*, M. Schillinger b, M. Mu¨llner a, T. Ho¨lzenbein c, K. Janata a, K. Bayegan a, A.N. Laggner a a
Department of Emergency Medicine, Vienna General Hospital, University of Vienna, Waehringer Guertel 18-20/6D, 1090 Vienna, Austria b Department of Angiology, Vienna General Hospital, University of Vienna, Waehringer Guertel 18-20/6D, 1090 Vienna, Austria c Department of Vascular Surgery, Vienna General Hospital, University of Vienna, Waehringer Guertel 18-20/6D, 1090 Vienna, Austria Received 18 April 2001; received in revised form 7 December 2001; accepted 13 December 2001
Abstract Background: There is increasing evidence that an inflammatory process is present in abdominal aortic aneurysms (AAAs) to varying degrees. The aim of this study was to compare acute phase reactants in patients with asymptomatic AAA, symptomatic AAA without rupture and ruptured AAA. Method: Two hundred and twenty-five consecutive patients treated because of AAA were included in this case-control study. Polynomial logistic regression analysis was applied to compare admission C-reactive protein (CRP) and white blood count (WBC) measured in 111 asymptomatic outpatients, 52 symptomatic patients without rupture and 62 patients with rupture of the aneurysm. We adjusted for the potentially confounding effect of age, sex, haemoglobin levels and aneurysm diameter. Results: Patients with symptomatic AAA and patients with ruptured AAA had significantly elevated CRP (p / 0.002) and WBC (p B/0.0001) levels compared to asymptomatic patients. There was no statistically significant difference in CRP and WBC between patients with symptomatic AAA and ruptured AAA. Median CRP values of asymptomatic, symptomatic and ruptured AAA were B/0.5 (interquartile range (IQR) B/0.5 /0.85), 1.1(IQR B/0.5 /4.0) and 2.4 mg/dl (IQR 0.65 /8.6), respectively, and median WBC values were 6.5 (IQR 5.5 /8.0), 8.7 (IQR 6.8 /11.2) and 13.2 (IQR 10.5 /17.0), respectively. Conclusion: A significant elevation of CRP and WBC could be found in patients who presented with symptoms or rupture of an AAA. These indicators of inflammation were not observed in asymptomatic patients with AAA. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Abdominal aortic aneurysm; Inflammation; CRP; Rupture
1. Introduction There is increasing evidence that inflammatory process is present in abdominal aortic aneurysms (AAAs) to varying degrees [1 /5]. Recent reports suggest that the inflammatory aortic aneurysm is not a distinct clinical entity and arises from the same pathogenic stimulus responsible to non-inflammatory AAAs [1]. A gradual passage in terms of inflammatory response from ordinary atherosclerotic to inflammatory AAA was described [4].
* Corresponding author. Tel.: 43-1-40-400-1964; fax: 43-1-40400-3953. E-mail address: [email protected] (H. Domanovits).
The degradation of the elastic media is a hallmark in the development of AAAs, a process which may be triggered by inflammatory response [6]. Acute phase reactants are expected to play a role in the progression of aortic disease [7]. However, the association of laboratory parameters of inflammation like C-reactive protein (CRP), white blood count (WBC) and fibrinogen with the clinical presentation and acuity of aortic aneurysm have not been investigated up to now. An increase of acute phase reactants in patients with ruptured AAA can be expected. However, signs of an underlying inflammatory process in symptomatic patients without rupture have not been evaluated. The aim of the study was to determine the association of acute phase reactants CRP, WBC and fibrinogen in patients with asymptomatic, symptomatic and ruptured AAA.
0021-9150/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 2 1 - 9 1 5 0 ( 0 2 ) 0 0 0 0 6 - 0
298
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
2. Methods 2.1. Patients Patients with symptomatic AAA without signs of rupture and patients with ruptured AAA were compared to patients with asymptomatic AAA. We included all consecutive patients with a symptomatic AAA admitted to the emergency department (ED) of a 2200 bed university hospital between 1 January, 1992 and 31 November, 1998. The comparison cohort comprised a group of consecutive patients with asymptomatic AAA treated at the outpatient unit of the Department of Vascular Surgery within the same time interval. These patients were scheduled for elective endovascular or surgical aneurysm repair. Patients with inflammatory aneurysm [8,9] based on computed tomography findings were not included. Concomitant inflammatory diseases which obviously cause elevation of acute phase reactants were excluded by clinical examination, chest radiography and urinalysis. Seven patients (6%) were not included in the final analysis because of coexistent inflammatory disease such as pneumonia (n /2), urinary tract infection (n /4) and cholecystitis (n /1). 2.2. Definitions An AAA was defined as a permanent dilatation of the aorta, with a diameter of at least 50% greater than the proximal aortic neck [10]. Rupture was defined by direct visualisation of discontinuity of the aortic wall by computed tomography, sonography, intraoperative findings or at autopsy. Symptomatic aneurysm was defined if a patient presented with pain, referred as lower back pain, abdominal or flank pain, which was supposed to be originating from the aneurysm. 2.3. Laboratory parameters Serum CRP level, WBC and plasma fibrinogen are routinely measured at the time of admission to the emergency department and during routine visits to the outpatient ward of the Department of Vascular Surgery. For measurement of serum CRP levels Tina-quant† (Roche) was used, WBC was measured by NE 9000† (Sysmex) and fibrinogen was measured by Fibrinogen Clauss† (Stago/Roche). 2.4. Follow up Patients who were treated conservatively were followed for 6 months for the occurrence of ruptureassociated death. These follow-up data were derived from the Austrian Central Statistical Office (Day and Cause of Death, ICD 9).
The study was performed according to the Declaration of Helsinki. 2.5. Statistical analysis Continuous data are presented as the median and the interquartile range (IQR, range from the 25th to 75th percentile). Percentages were determined for dichotomous variables. For univariate comparison of age, sex haemoglobin and aneurysm diameter between the groups the x2-test and the Kruskal Wallis test was used. Spearman’s rank correlation coefficient was calculated to assess the strength of association between continuous variables. Polynomial logistic regression models were applied to assess the independent association of CRP, WBC and fibrinogen between the three groups (symptomatic AAA vs. ruptured AAA vs. asymptomatic AAA) and to adjust for patients, age, sex and haemoglobin levels. A P -value 5/0.05 was accepted as statistically significant. All calculations were performed with Excel for Windows, and SPSS (Version 10.0) for Windows.
3. Results Within the observation period 114 eligible patients presented at the ED because of symptomatic AAA. In 62 of these patients rupture of the AAA was diagnosed. The median aneurysm diameter in patients with ruptured aneurysm was 80 mm (IQR 70 /90), the median aneurysm diameter in patients with symptomatic aneurysms without rupture was 60 mm (IQR 50/70). The comparison group comprised 111 patients from the outpatient ward of the Department of Vascular Surgery with asymptomatic AAA with a median aneurysm diameter of 54 mm (IQR 50 /63) (P B/0.0001). Median age of all patients was 72 years (IQR 66 /76) and 187 patients (83%) were male. Baseline data and laboratory findings of the three groups are presented in Table 1. Age was similar in the three groups (P /0.7), the sex distribution differed significantly (P /0.009). Haemoglobin was also significantly different, patients with ruptured AAA had the lowest levels (P B/0.0001). Differences of baseline CRP and WBC levels were found between the three groups using univariate logistic regression (P B/0.0001) (Table 2). CRP and WBC were both taken as an indicator for acute phase reaction, but were not analysed in the same multivariate model for reasons of collinearity. Two separate multivariate polynomial logistic regression models were applied to assess the independent association of CRP and WBC with the group classification of AAA and to adjust for patients age, sex, haemoglobin levels and aneurysm diameter. Compared to asymptomatic patients, CRP and WBC were independently
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
299
Table 1 Baseline data of 225 patients with AAA
Median age in years (IQR) Male sex (%) Median haemoglobin (g/dl) (IQR) Median CRP (mg/dl) (IQR) WBC per 1000 (IQR) Fibrinogen (mg/dl) (IQR)
Asymptomatic AAA (n 111)
Symptomatic AAA (n 52)
Ruptured AAA (n 62)
73 (67 /76) 100 (90%)* 14.3 (13.2 /15.2)* B 0.5 ( B 0.5 /0.85)* 6.5 (5.5 /8.0)* 327 (273 /392)
72 (67 /77) 37 (71%)* 12.2 (11.1 /14.5)* 1.1 ( B 0.5 /4.0)* 8.7 (6.8 /11.2)* 388 (308 /466)
72 (65 /77) 50 (81%)* 10.3 (9.2 /12.6)* 2.4 (0.65 /8.6)* 13.2 (10.5 /17.0)* 309 (216 /433)*
AAA, abdominal aortic aneurysm; IQR, interquartile range; CRP, C-reactive protein; WBC, white blood count. * P B 0.05.
elevated in patients with symptomatic disease and in patients with ruptured AAA after adjusting for age and sex. When additionally adjusting for haemoglobin and aneurysm diameter, minimal changes of odds ratios were observed. The difference of CRP and WBC between the three groups, however, remained highly significant (Tables 3 /5). When additionally adjusting for aneurysm diameter the odds ratios and the 95% confidence interval (CI) remained largely unchanged. Therefore, it seems unlikely that CRP merely reflects the aneurysm size. Furthermore, in symptomatic patients, there was no correlation between the aneurysm diameter and the CRP level (Spearman’s rank correlation r / 0.06, P /0.8). To answer the question of the relative contribution of low haemoglobin vs. high CRP in predicting those with symptoms, we tested for interaction between these variables in the polynomial model: when CRP and haemoglobin were entered into the model with and without a multiplicative interaction term, no significant alteration of the model fit was observed, thus we believe that CRP does not vary over levels of haemoglobin. Table 2 Association of serum CRP levels, WBC and fibrinogen with the clinical presentation of AAA in univariate polynomial logistic regression models: asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA OR
95% CI
P -value
CRP Asymptomatic AAA (n 111) Symptomatic AAA (n 52) Ruptured AAA (n 62)
1.0 1.3 1.4
/ 1.2 /1.6 1.2 /1.6
/ B 0.0001 B 0.0001
WBC Asymptomatic AAA (n 111) Symptomatic AAA (n 52) Ruptured AAA (n 62)
1.0 1.4 1.8
/ 1.3 /1.7 1.5 /2.1
/ B 0.0001 B 0.0001
Fibrinogen Asymptomatic AAA (n 111) Symptomatic AAA (n 52) Ruptured AAA (n 62)
1.0 1.0 1.0
/ 1.0 /1.0 1.0 /1.0
/ 0.05 0.6
AAA, abdominal aortic aneurysm; IQR, interquartile range; CRP, Creactive protein; WBC, white blood count; OR, odds ratio; CI, confidence interval; /, denotes reference group.
We applied a similar model using the variable fibrinogen as an indicator for the acute phase reaction, which showed no significant differences between the three groups (data not shown) (data presented in the appendix for information of editors and reviewers). Fifty-two patients with initially conservative treatment were followed for up to 6 months to investigate the Table 3 Independent association of serum CRP levels with asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA in a multivariate polynomial logistic regression model OR
95% CI
P -value
Model controlled for age (years ) and sex Asymptomatic AAA (n 111) 1.0 Symptomatic AAA (n 52) 1.3 Ruptured AAA (n 62) 1.4
/ 1.2 /1.5 1.2 /1.6
/ B 0.0001 B 0.0001
Model controlled for age (years ), sex , haemoglobin (g /dl ) and aneurysm diameter (cm ) Asymptomatic AAA (n 111) 1.0 / / Symptomatic AAA (n 52) 1.3 1.1 /1.5 0.008 0.007 Ruptured AAA (n 62) 1.3 1.1 /1.5 AAA, abdominal aortic aneurysm; IQR, interquartile range; OR, odds ratio; CI, confidence interval; /, denotes reference group.
Table 4 Independent association of WBC with asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA in a multivariate polynomial logistic regression model OR
95% CI
P -value
Model controlled for age (years ) and sex Asymptomatic AAA (n 111) 1.0 Symptomatic AAA (n 52) 1.5 Ruptured AAA (n 62) 1.8
/ 1.3 /1.7 1.5 /2.1
/ B 0.0001 B 0.0001
Model controlled for age (years ), sex , haemoglobin (g /dl ) and aneurysm diameter (cm ) Asymptomatic AAA (n 111) 1.0 / / Symptomatic AAA (n 52) 1.8 1.4 /2.2 B 0.0001 B 0.0001 Ruptured AAA (n 62) 2.0 1.5 /2.5 AAA, abdominal aortic aneurysin; IQR, interquartile range; OR, odds ratio; CI, confidence interval; /, denotes reference group.
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
300
Table 5 Independent association of fibrinogen with asymptomatic AAA (reference group) vs. symptomatic AAA vs. ruptured AAA in a multivariate polynomial logistic regression model OR
95% CI
P -value
Model controlled for age (years ) and sex Asymptomatic AAA (n 111) 1.0 Symptomatic AAA (n 52) 1.0 Ruptured AAA (n 62) 1.0
/ 1.0 /1.0 1.0 /1.0
/ 0.04 0.9
Model controlled for age (years ), sex , haemoglobin (g /dl ) aneurysm diameter (cm ) Asymptomatic AAA (n 111) 1.0 / Symptomatic AAA (n 52) 1.0 1.0 /1.0 Ruptured AAA (n 62) 1.0 1.0 /1.0
and / 0.1 0.4
AAA, abdominal aortic aneurysm; IQR, interquartile range; OR, odds ratio; CI, confidence interval, /, denotes reference group.
association of baseline CRP and WBC levels and the occurrence of death due to aneurysm rupture. Eleven patients died due to aneurysm rupture during this time interval. These patients had significantly higher baseline CRP levels (median 3.9 mg/dl, IQR 1.6 /13.3) compared to patients without death due to rupture (median 0.9 mg/dl, IQR B/0.5 /1.8) (P /0.006). Furthermore, patients with death due to rupture had significantly higher baseline WBC (median 10.4/1000, IQR 8.8 /14.1) compared to survivors without rupture (median 7.6/1000, IQR 6.2 /10.5) (P /0.004).
4. Discussion We found higher serum CRP and WBC levels in patients with symptomatic and ruptured AAA compared to asymptomatic patients. Not surprisingly elevated laboratory parameters of inflammation were present in patients with ruptured aneurysm. Remarkably, there was also a significant elevation in symptomatic patients without rupture. Others have shown aneurysmal pain to be associated with expansion or wall distension with impending rupture or dissection [11 /13]. We found aneurysmal pain associated with an elevation of CRP and WBC. However, we could not document expansion or wall distension due to the cross sectional design of our study. As aneurysmal pain was associated with elevated CRP and WBC, our findings suggest, that an inflammatory process is present in the aortic wall which promotes expansion and rupture. This inflammatory process may be quantified by measurement of serum CRP and WBC. 4.1. Inflammatory activation Inflammatory activation is an important pathophysiologic process in several atherosclerotic entities like
coronary artery disease, peripheral artery disease and aortic aneurysm [1 /4,9,14/17]. Inflammatory cytokines were shown to be elevated in patients with AAA [18], which lead to the degradation of extracellular matrix components [19,20], a basic feature in the pathogenesis of AAA. The cause of the inflammatory process in the aneurysmal wall is unknown; it may be an autoimmune reaction as well as an infectious agent like chlamydia pneumoniae [21]. The elevation of CRP, WBC and fibrinogen may be an indicator of acute phase response due to vascular tissue damage and inflammation in atherosclerotic disease [22 /24]. However, so far it cannot be distinguished to what degree the elevated serum markers of inflammation in patients with acute aortic aneurysm are due to local vascular inflammatory process or due to acute phase reaction accompanying the systemic inflammatory response syndrome.
4.2. Diagnostic value The diagnostic value of CRP and WBC in patients with AAA has not been evaluated. Concomitant disorders with overlapping signs and symptoms occur frequently in the elderly population with AAA [25,26]. Abdominal, flank or back pain are the most common symptoms in patients with an expanding or ruptured AAA [11,12]. However, patients with expanding AAA may also present with a paucity of symptoms and even with ruptured AAA patients may fail to manifest signs of hemodynamic instability upon rupture [27,28]. Signs and clinical symptoms in patients with AAA are sometimes not a reliable indicator for the acuity of the disease. Acute phase reaction as a measurable parameter might be useful in differentiation of patients with and without acute disease.
4.3. Prognostic value The prognostic value of elevated CRP and WBC in patients with symptomatic AAA may be discussed: during up to 6 months follow-up period, 11 out of 52 patients with symptomatic AAA without rupture died due to rupture of the aneurysm. Significantly higher levels of CRP and WBC were found at presentation in the symptomatic non-survivors compared to symptomatic survivors. However, larger patient samples who are primarily scheduled for conservative treatment will be necessary to confirm these findings, since the majority of patients included in the present study was operated upon the aneurysm during the follow-up period and therefore was not followed for aneurysm rupture or death.
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302
4.4. Therapeutic implication Chronic inflammation indicated by laboratory parameters may play a role in the progression of aortic disease [7]. Experimental approaches with anti-inflammatory therapy successfully inhibited the progress of disease [29 /31]. Particularly, the control of inflammatory response via suppression of proteolytic enzymes showed promising results in animal models [30,31]. Comparable data in humans are not available. Accordingly, the success of different conservative therapeutic strategies might be measured by monitoring the course of inflammation markers. 4.5. Limitations For measurement of serum CRP levels a standard CRP assay was used. High sensitivity CRP assays are required for atherosclerotic risk prediction in apparently healthy individuals [32]. Patients in this study with symptomatic disease had median baseline CRP levels of 1.1 and 2.4 mg/dl, which is significantly above the detection level (0.3 mg/dl) of the used assay. Therefore, the standard assay seems to be appropriate for this setting. 4.6. Conclusion An increase of CRP and WBC could be found in patients with symptomatic AAA. Remarkably, even in symptomatic patients without ruptures significant acute phase reaction was observed.
Acknowledgements The authors thank Professor Dr E. Schuster from the Institute for Medical Computer Sciences for statistical advice.
References [1] Rasmussen TE, Hallett JW. New insights into inflammatory aortic aneurysms. Eur J Vasc Endovasc Surg 1997;14:329 /32. [2] Rose AG, Dent DM. Inflammatory variant of abdominal aortic atherosclerotic aneurysm. Arch Pathol Lab Med 1981;105:409 / 13. [3] Pennel RC, Hollier LH, Lie JT. Inflammatory abdominal aortic aneurysms: a thirty year review. J Vasc Surg 1985;2:859 /69. [4] Sterpetti AV, Hunter WJ, Feldhaus RJ. Inflammatory aneurysm of the abdominal aorta: incidence, pathologic, and etiologic considerations. J Vasc Surg 1989;9:643 /50. [5] Rehm JP, Grange JJ, Baxter BT. The formation of aneurysms. Semin Vasc Surg 1998;11:193 /202.
301
[6] Davis V, Persidskaia R, Baca Regen L, Itoh Y, Nagase H, Persidsky Y, et al. Matrix metalloproteinase-2 production and its binding to the matrix are increased in abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol 1998;18:1625 /33. [7] Rohde LE, Arroyo LH, Rifai N, Creager MA, Libby P, Ridker PM, et al. Plasma concentrations of interleukin-6 and abdominal aortic diameter among subjects without aortic dilatation. Arterioscler Thromb Vasc Biol 1999;19:1695 /9. [8] Walker DI, Bloor K, Williams G, Gillie I. Inflammatory aneurysms of the abdominal aorta. Br J Surg 1972;59:609 /14. [9] Rasmussen TE, Hallett JW, Jr.. Inflammatory abdominal aortic aneurysms: a review with new perspectives in etiology. Ann Surg 1997;225:1 /10. [10] Johnston KW, Rutherford RB, Tilson MD, Shah DM, Hollier L, Stanley JC. Suggested standard for reporting on arterial aneurysms. J Vasc Surg 1991;13:452 /8. [11] Vohra RD, Groome J. Long-term survival in patients undergoing resection of abdominal aortic aneurysm. Ann Vasc Surg 1990;4:460 /5. [12] Darke SG, Eadie DDG. Abdominal aortic aneurysmectomy: a review of 60 consecutive cases contrasting elective and emergency surgery. J Cardiovasc Surg 1973;14:484 /91. [13] Sullivan CA, Rohrer MJ, Cutler BS. Clinical management of the symptomatic but unruptured abdominal aortic aneurysm. J Vasc Surg 1990;11:799 /803. [14] Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. Plasma concentration of C-reactive protein and risk of developing peripheral vascular disease. Circulation 1998;97:425 / 8. [15] Heinrich J, Schulte H, Schoenfeld R, Koehler E, Assmann G. Association of variables of coagulation, fibrinolysis and acutephase with atherosclerosis in coronary and peripheral arteries and those arteries supplying the brain. Thromb Haemostasis 1995;73:374 /9. [16] Liuzzo G, Biasucci L, Gallimore JR, Grillo RL, Rebuzzi AG, Pepys MB, et al. The prognostic value of C-reactive protein and serum amyloid A protein in severe unstable angina. N Engl J Med 1994;331:417 /24. [17] Mendall MA, Patel P, Ballam L, Strachan D, Northfield TC. C reactive protein and its relation to cardiovascular risk factors: a population based cross sectional study. Br Med J 1996;312:1061 / 5. [18] Juvonen J, Surcel HM, Satta J, Teppo AM, Bloigu A, Syrjala H, et al. Elevated circulating levels of inflammatory cytokines in patients with abdominal aortic aneurysm. Atheroscler Thromb Vasc Biol 1997;17:2843 /7. [19] Satta J, Laurila A, Paakko P, Haukipuro K, Sormunen R, Parkkila S, et al. Chronic inflammation and elastin degradation in abdominal aortic aneurysm disease: an immunohistochemical and electron microscopic study. Eur J Vasc Endovasc Surg 1998;15:313 /9. [20] Grange JJ, Davis V, Baxter BT. Pathogenesis of abdominal aortic aneurysm: an update and look toward the future. Cardiovasc Surg 1997;5:256 /65. [21] Lindholt JS, Fasting H, Henneberg EW, Ostergaard L. A review of chlamydia pneumoniae and atherosclerosis. Eur J Vasc Endovasc Surg 1999;17:283 /9. [22] Danesh J, Collins R, Appleby P, Peto R. Fibrinogen, C-reactive protein, albumin or white cell count: meta-analyses of prospective studies of coronary heart disease. J Am Med Assoc 1998;279:1477 /82. [23] Ridker PM. C-reactive protein and risks of future myocardial infarction and thrombotic stroke. Eur Heart J 1998;19:1 /3. [24] Maseri A, Biasucci LM, Liuzzo G. Inflammation in ischemic heart disease. Br Med J 1996;312:1049 /50. [25] Rothrock SG, Greenfield RH, Falk JL. Acute abdominal emergencies in the elderly: clues to identifying serious illness.
302
[26]
[27]
[28]
[29]
H. Domanovits et al. / Atherosclerosis 163 (2002) 297 /302 Part I */clinical assessment and diagnostic studies. Emerg Med Rep 1992;13:177 /84. Rothrock SG, Greenfield RH, Falk JL. Acute abdominal emergencies in the elderly: clinical evaluation and management. Part II */diagnosis and management of common conditions. Emerg Med Rep 1992;13:185 /92. Zimmer T. Absence of back pain and tachycardia in the emergency presentation of abdominal aortic aneurysm. Am J Emerg Med 1988;6:316 /21. Bower TC, Cherry KJ, Pairolero PC. Unusual manifestations of abdominal aortic aneurysms. Surg Clin North Am 1989;69:745 / 54. Miralles M, Wester W, Sicard GA, Thompson R, Reilly JM. Indomethacin inhibits expansion of experimental aortic aneurysm
via inhibition of the cox2 isoform of cyclooxygenase. J Vasc Surg 1999;29:884 /92. [30] Curci JA, Petrinec D, Liao S, Golub LM, Thompson RW. Pharmacologic suppression of experimental aortic aneurysms: a comparison of doxycycline and four chemically modified tetracyclines. J Vasc Surg 1998;28:1082 /93. [31] Bigatel DA, Elmore JR, Carey DJ, Cizmeci Smith G, Franklin DP, Youkey JR. The matrix metalloproteinase inhibitor BB-94 limits expansion of experimental abdominal aortic aneurysms. J Vasc Surg 1999;29:130 /8. [32] Roberts WL, Sedrick R, Moulton L, Spencer A, Rifai N. Evaluation of four automated high-sensitivity C-reactive protein methods: implications for clinical and epidemiological applications. Clin Chem 2000;46:461 /8.