Professional Documents
Culture Documents
Background. A systemic inflammatory response is not Results. The prevalence of AA, AG, and GG TNF-308
uncommonly observed after coronary revascularization. genotype was 12%, 40%, and 48%, respectively. Patients
Tumor necrosis factor alpha is one of a number of modu- homozygous for A had higher circulating levels of TNF
lators of this response. A functional polymorphism within (p 0.009). Higher levels of TNF were significantly
the TNF gene at position G-308A has been associated with associated with prolonged intensive care unit stay (p
increased TNF levels. The relationship between predicted 0.008), increase usage of an inotropic agent (p 0.024),
TNF genotype and circulating TNF levels in patients increased mortality risk (p 0.018), and diabetes (p
undergoing coronary revascularization surgery has yet to be 0.019). These remained statistically significant after risk
defined. We examined the relationship between TNF stratification.
G-308A polymorphism, TNF postoperative levels, and Conclusions. Patients of the AA-308 TNF genotype
clinical outcome after coronary revascularization surgery. showed significantly higher TNF plasma levels. Higher
Methods. We obtained DNA from 96 consecutive pa- plasma levels of TNF were associated with less favor-
tients who underwent elective coronary revascularization. able outcome after coronary revascularization surgery. It
Patients were genotyped for TNF G-308A polymorphism may prove useful to utilize TNF serum levels as a
using sequence specific primerpolymerase chain reaction marker for identifying high-risk patients in the future.
(SSP-PCR). Tumor necrosis factor alpha levels were mea-
sured on serum samples taken 3 hours postoperatively (Ann Thorac Surg 2006;81:132 8)
using enzyme-linked immunosorbent assay (ELISA). 2006 by The Society of Thoracic Surgeons
CARDIOVASCULAR
2006;81:132 8 TNF RESPONSE AFTER CORONARY SURGERY
monary bypass was instituted in 82 patients by cannula- Table 1. Primer Sequences Used for the TNF Sequence-
tion of the right atrium and ascending aorta. Myocardial Specific Primer Genotyping Method
protection was provided by intermittent antegrade with Primer Position Primer Sequence Tm
or without retrograde blood cardioplegia. Aprotinin was
not used. In 14 patients, the coronary revascularization TNF-308 A 5=-ATAGGTTTTGAGGGGCATGA-3= 60C
was performed without the use of CPB with the aid of TNF-308 G 5=-ATAGGTTTTGAGGGGCATGG-3=
intracoronary shunt (Intravascular Arteriotomy Cannula; Generic 5=-TCTCGGTTTCTTCTCCATCG-3=
Medtronic, Minneapolis, MN) and stabilizer (Medtronic antisense
Octopus 2 Tissue Stabilization System). Perioperative HGH (sense) =5-GCCTTCCCAACCATTCCCTTA-3= 60C
anticoagulation with heparin was reversed after CPB HGH (antisense) =5-TCAGGATTTCTGTTGTGTTTC-3=
with the use of protamine sulphate. HGH human growth hormone; Tm melting temperature;
TNF tumornecrosis factor.
Sample Collection
Venous blood samples were collected preoperatively in
the outpatient clinic and 3 hours after surgery in the each for dATP, dCTP, dGTP, and Dttp (all ABgene,
intensive care unit in ethylenediamine-tetra acid (EDTA) Epsom, United Kingdom). Precipitant and red dye were
Vacutainer tubes (Becton Dickinson, Plymouth, UK). also present for electrophoresis. This reaction mixture
Samples were centrifuged within 30 minutes of collection was added to human growth hormone primers (control)
at 3,000 rpm for 10 minutes. Plasma and blood cells were in a ratio of 10 l:1 l. Ten microliters of this was then
separated and stored separately at 80C. mixed with 1.5 l DNA in a thin-walled PCR tube; 5 l of
this mastermix was added to 5 l of specific G (5-
Genotyping ATAGGTTTTGAGGGGCATGG-3)or A (5-
EXTRACTION OF DNA. The DNA was obtained from EDTA ATAGGTTTTGAGGGGCATGA-3) primers in separate
anticoagulated blood using the double lysis method. tubes. Cycling conditions included an initial denatur-
Briefly, 4 mL collected blood was centrifuged; blood cells ation of 95C for 1 minute followed by 95C for 15
were transferred to 13 mL polypropylene tubes contain- seconds, 65C for 50 seconds, and 72C for 40 seconds (10
ing 9 mL Lysis buffer 1 (155 mM ammonium chloride, 10 cycles), followed by 95C for 50 seconds, 59C for 50
mM potassium hydrogen carbonate, 1 mM EDTA) and seconds, and 72C for 50 seconds (20 cycles). A final
mixed for 15 minutes. Samples were centrifuged and holding temperature at 4C was performed. The PCR
supernatants removed. The cell pellets were then resus- products were electrophoresed on a 2% agarose gel (Life
pended and lysed again. A nuclear membrane lysis step Technologies) and visualized using ethidium bromide
was performed using 25 mM EDTA and 2% SDS in a 3 (Life Technologies) under ultraviolet illumination [7, 8].
mL volume and 1 mL 10M ammonium acetate. Samples Enzyme-Linked Immunosorbent Assay
were centrifuged and supernatant added to propan-2-ol
Serum levels of TNF were measured using a solid phase
to wash DNA. This was finally centrifuged and resus-
sandwich enzyme-linked immunosorbent assay (ELISA,
pended in 350 l double-distilled water.
IDS Ltd, Tyne and Wear, UK). A monoclonal TNF
SEQUENCE-SPECIFIC PRIMERPOLYMERASE REACTION. The poly-
antibody was coated onto the wells of a 96-well plate.
merase chain reaction is a method of gene amplification. Standards of known TNF concentration were then
Polymerase chain reaction amplifies the chosen DNA in added along with control samples and patient serum.
the test tube. Two short single-stranded DNA fragments, After washing, a biotinylated polyclonal TNF antibody
or primers, delineate the segment to be amplified. The was added and incubated. After washing, streptavidin-
primers initiate the amplification, which proceeds in coupled peroxidase was added. After incubation, a wash
successive copying cycles, each of which double the step was performed to remove a chromogenic peroxi-
number of DNA segments in the reaction. A cycle begins dase. Finally, tetramethylbenzidine (TMB) substrate was
when heat melts the double-stranded DNA template into added that changed color directly in proportion to the
single strands. The primers hybridize to their comple- amount of TNF present. The color (absorbance) was then
mentary sequences on the separated strands of the measured using a Dynex Technologies MRX plate reader
template. This process can be repeated until it incorpo- (DYNEX Technologies Ltd, Worthing, West Sussex, UK)
rates all the primers into double-stranded DNA. It is with a primary wavelength of 450 nm and a reference
possible to make millions of copies of a DNA segment in wavelength of 620 nm.
a matter of hours with PCR.
The DNA was amplified using specific oligonucleotide Statistical Analysis
primers based on the published sequence (GenBank The 2 test was used to analyze relationship between
accession number AF 005485, Genosys, Suffolk, UK). The categorical data. Nonparametric Mann-Whitney U test
primer sequences used for TNF genotyping are shown and Kruskal-Wallis test, as appropriate, were used to
in Table 1. compare TNF levels and different outcomes between
Each PCR reaction mixture comprised 1.25 U Thermo- subgroups. Nonparametric testing was chosen because
prime Plus DNA polymerase, 75 mM Tris-HCl, 20 mM (apart from age, which was analyzed using one-way
(NH4) 2SO4, 2.0 mM MgCl2, 0.01% Tween 20, and 0.2 mM analysis of variance) the data did not follow a normal
134 BITTAR ET AL Ann Thorac Surg
CARDIOVASCULAR
Fig 1. Tumor necrosis factor alpha genotypes prevalence. (Black bars Fig 3. Tumor necrosis factor alpha (TNF) postoperative serum levels
normal population; gray bars study patients.) and intensive care unit stay (p 0.008).
Ann Thorac Surg BITTAR ET AL 135
CARDIOVASCULAR
2006;81:132 8 TNF RESPONSE AFTER CORONARY SURGERY
tively, which was not different from normal population as ity, and granuloma development. Excessive production
predicted by Hardy-Weinberg equilibrium (Fig 1). Pa- of TNF may lead to organ dysfunction and death [13].
tients with AA genotypes had higher circulating levels of Tumor necrosis factor alpha acts by binding to specific
TNF postoperatively compared with AG and GG geno- receptors on cell surfaces, blocking these receptors pro-
type, which reached high statistical significance TNF vide a protective mechanism. Anti-TNF therapy has
levels: AA 511 pg/mL (range, 30 to 4,985), AG 38 become a well-recognized treatment modality in several
pg/mL (range, 25 to 3,309), and GG 37 pg/mL (range, 25 inflammatory conditions such as Crohns disease and
to 3,512; p 0.009). These results remained statistically rheumatoid arthritis [14, 15].
significant after risk adjustment for age, sex, blood loss, In patients undergoing coronary revascularization sur-
ventilation time, duration of CPB, and aortic cross-clamp gery, different stimuli such as general anesthesia, surgi-
time (Fig 2, Tables 3 and 4). cal wounds, heparin administration, CPB, and protamine
Higher levels of TNF were associated with prolonged administration are thought to play a role in the genesis of
intensive care unit stay (Kruskal-Wallis test, p 0.008), this response [16]. Previous reports confirmed the increase
increased usage of inotropes (Mann-Whitney test, p of TNF levels in response to cardiac surgery at different
0.024), and higher incidence of death (Mann-Whitnet time points reaching a peak at 3 hours after surgery and
test, p 0.018), Diabetic patients had significantly higher degraded rapidly with a short half-life [3, 4, 17].
levels of TNF compared with nondiabetic patients (p This study has focused on the aspects of the relation-
0.019). These results remained statistically significant ship between TNF genotype, TNF postoperative lev-
after risk adjustment to age, sex, aortic cross-clamp time, els, and postoperative outcome. We hypothesized that
and duration of CPB using multiple linear regression polymorphisms in the TNF gene -308 may account for
analyses (Figs 3 through 6). the variation observed in TNF plasma concentrations
after cardiac revascularization. Our results showed a
highly significant positive association between TNF
Comment levels and AA genotype. The distribution of TNF
Inflammation is involved in all stages of atherosclerotic G-308A genotypes in this cohort were close to the normal
development. It is a cause and a consequence of ischemic distribution predicted by the Hardy-Weinberg equilib-
heart disease [9 11]. Tumor necrosis factor alpha is a rium. Although we found a significant association be-
proinflammatory mediator in the pathogenesis of the tween the AA genotype and the postoperative TNF
SIRS [12]. It has been previously demonstrated that TNF levels, we were unable to demonstrate a significant
plays a major role in SIRS secondary to infection, burns, association between TNF genotype and clinical out-
trauma, hemorrhagic shock, and pancreatitis [5]. Tumor come. This could be related to the small number of
necrosis factor alpha influences the outcome of other patients with the AA genotype in the study group.
inflammatory processes, including allograft rejection, We examined the relationship between TNF postop-
ischemia-reperfusion injury, delayed-type hypersensitiv- erative levels and clinical outcome, and we found that
References
1. Rothenburger M, Soeparwata R, Deng MC, et al. Prediction
of clinical outcome after cardiac surgery: the role of cyto-
kines, endotoxin, and anti-endotoxin core antibodies. Shock
2001;16(Suppl 1):44 50.
2. Butler J, Baigrie RJ, Parker D, et al. Systemic inflammatory
Fig 5. Tumor necrosis factor alpha (TNF) postoperative serum levels responses to cardiopulmonary bypass: a pilot study of the
and mortality in the study group population (p 0.018). effects of pentoxifylline. Respir Med 1993;87:285 8.
Ann Thorac Surg BITTAR ET AL 137
CARDIOVASCULAR
2006;81:132 8 TNF RESPONSE AFTER CORONARY SURGERY
3. Misoph M, Babin-Ebell J. Interindividual variations in cyto- cell death on rheumatoid arthritis (RA) synovial cells: a
kine levels following cardiopulmonary bypass. Heart Vessels possible mechanism of rheumatoid synovial hyperplasia
1997;12:119 27. and a clinical benefit of anti-TNF-alpha therapy for RA.
4. Wei M, Kuukasjarvi P, Laurikka J, et al. Cytokine responses Cytokine 2000;12:281 8.
in patients undergoing coronary artery bypass surgery after 16. Liebold A, Keyl C, Birnbaum DE. The heart produces but the
ischemic preconditioning. Scand Cardiovasc J 2001;35:142 6. lungs consume proinflammatory cytokines following cardio-
5. Pannen BH, Robotham JL. The acute-phase response. New pulmonary bypass. Eur J Cardiothorac Surg 1999;15:340 5.
Horiz 1995;3:18397. 17. Tonnesen E, Christensen VB, Toft P. The role of cytokines in
6. Tomasdottir H, Hjartarson H, Ricksten A, Wasslavik C, cardiac surgery. Int J Cardiol 1996;53(Suppl):110.
Bengtsson A, Ricksten SE. Tumor necrosis factor gene poly- 18. Majetschak M, Flohe S, Obertacke U, et al. Relation of a TNF
morphism is associated with enhanced systemic inflamma-
gene polymorphism to severe sepsis in trauma patients. Ann
tory response and increased cardiopulmonary morbidity
Surg 1999;230:20714.
after cardiac surgery. Anesth Analg 2003;97:944 9 [and Table
of Contents]. 19. Antoniades C, Tousoulis D, Tountas C, et al. Vascular endo-
7. Aldener-Cannava A, Olerup O. HLA-DOB1 low-resolution thelium and inflammatory process, in patients with combined
typing by PCR amplification with sequence-specific primers type 2 diabetes mellitus and coronary atherosclerosis: the
(PCR-SSP). Eur J Immunogenet 1994;21:44755. effects of vitamin C. Diabet Med 2004;21:552 8.
8. Perrey C, Turner SJ, Pravica V, Howell WM, Hutchinson IV. 20. Kubaszek A, Pihlajamaki J, Komarovski V, et al. Promoter
ARMS-PCR methodologies to determine IL-10, TNF-alpha, polymorphisms of the TNF-alpha (G-308A) and IL-6 (C-
TNF-beta and TGF-beta 1 gene polymorphisms. Transpl 174G) genes predict the conversion from impaired glucose
Immunol 1999;7:127 8. tolerance to type 2 diabetes: the Finnish Diabetes Prevention
9. Ridker PM, Rifai N, Stampfer MJ, Hennekens CH. Plasma Study. Diabetes 2003;52:1872 6.
concentration of interleukin-6 and the risk of future myocar- 21. Fernandez-Real JM, Lainez B, Vendrell J, et al. Shedding of
dial infarction among apparently healthy men. Circulation TNF-alpha receptors, blood pressure, and insulin sensitivity
2000;101:176772. in type 2 diabetes mellitus. Am J Physiol Endocrinol Metab
10. Ridker PM, Hennekens CH, Buring JE, Rifai N. C-reactive 2002;282:E9529.
protein and other markers of inflammation in the prediction 22. Yu BJ, Li JS, Zhang DL, Tang XM. The associations of the
of cardiovascular disease in women. N Engl J Med 2000;342: single nucleotide polymorphisms on TNF and CD14 promot-
836 43. ers with the mortality of infection, systematic inflammatory
11. Jenny NS, Tracy RP, Ogg MS, et al. In the elderly, interleu- response syndromec and sepsis in surgical patients [trans-
kin-6 plasma levels and the -174GC polymorphism are lation]. Zhonghua Yi Xue Za Zhi 2003;83:2132 6.
associated with the development of cardiovascular disease.
23. Gaudino M, Di Castelnuovo A, Zamparelli R, et al. Genetic
Arterioscler Thromb Vasc Biol 2002;22:2066 71.
control of postoperative systemic inflammatory reaction and
12. Cain BS, Meldrum DR, Dinarello CA, Meng X, Banerjee A,
Harken AH. Adenosine reduces cardiac TNF-alpha produc- pulmonary and renal complications after coronary artery
tion and human myocardial injury following ischemia- surgery. J Thorac Cardiovasc Surg 2003;126:110712.
reperfusion. J Surg Res 1998;76:11723. 24. Bittar MN, Carey JA, Barnard J, et al. Interleukin-6 G-174C
13. Strieter RM, Kunkel SL, Bone RC. Role of tumor necrosis polymorphism influences outcome following coronary re-
factor-alpha in disease states and inflammation. Crit Care vascularisation surgery. Heart Surg Forum 2005;8:E140 5.
Med 1993;21:S447 63. 25. Gaudino M, Andreotti F, Zamparelli R, et al. The -174G/C
14. Casellas i Jorda F. TNF-alpha inhibitors in inflammatory interleukin-6 polymorphism influences postoperative inter-
bowel disease [translation]. Med Clin (Barc) 2004;123:62734. leukin-6 levels and postoperative atrial fibrillation. Is atrial
15. Ohshima S, Mima T, Sasai M, et al. Tumour necrosis factor fibrillation an inflammatory complication? Circulation 2003;
alpha (TNF-alpha) interferes with Fas-mediated apoptotic 108(Suppl 1):1959.
INVITED COMMENTARY
Bittar and colleagues discuss whether the inflammatory with molecular mechanisms of the systemic inflamma-
response (particularly after coronary bypass operations) tory response. However, the systemic inflammatory re-
can be predicted by genetic phenotyping [1]. sponse involves a multitude of humeral and cellular
The bypass circuit (ie, blood foreign surface interface) mechanisms, and it would be an oversimplification to
causes an intense pro-inflammatory reaction. This re- assume that one molecule is responsible for this very
sponse seems to be macrophage and endothelial-cell complex phenomenon.
induced. The effect of tumor necrosis factor (TNF) by Genetic polymorphism is defined as the presence of
itself seems to be negative inotropic and pro- multiple alleles at a gene locus in appreciable frequen-
inflammatory. Some cytoprotective effects have been cies in a population. The TNF gene is embedded within
proposed as well. TNF acts pro-inflammatory by self the class three major histocompatibility complex on the
amplification as well as by induction and expression of short arm of chromosome 6. Whether TNF gene poly-
other pro-inflammatory cytokines. The myocardium con- morphism influences TNF expression in patients under-
stitutively expresses a low level of TNF. Although mac- going coronary artery bypass grafting with or without
rophages are activated at the blood foreign surface inter- cardiopulmonary bypass has been the object of a number
face, an increase of myocardial TNF after ischemia and of studies. The results have been conflicting.
reperfusion has also been demonstrated. It is unclear if How can we explain these discrepancies? First when
TNF gene expression and peptide synthesis occur dur- planning or reading one of these studies, the goal at the
ing global ischemia. Hence, the source of TNF is mul- onset has to be kept in mind. To study gene polymor-
tifactorial. It is intriguing to merge molecular genetics phism and its possible influence on outcome as a conse-