Cardiac Troponins I and T Are Biological Markers of Left Ventricular Dysfunction in Septic Shock

of 8

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
PDF
8 pages
0 downs
4 views
Share
Description
Cardiac Troponins I and T Are Biological Markers of Left Ventricular Dysfunction in Septic Shock
Tags
Transcript
  Cardiac Troponins I and T Are Biological Markersof Left Ventricular Dysfunction in Septic Shock  Kristien M. ver Elst, 1 Herbert D. Spapen, 2 Duc Nam Nguyen, 2 Christian Garbar, 3 Luc P. Huyghens, 2 and  Frans K. Gorus 1* Background:  Cardiac depression in severe sepsis andseptic shock is characterized by left ventricular (LV)failure. To date, it is unclear whether clinically unrec-ognized myocardial cell injury accompanies, causes, orresults from this decreased cardiac performance. Wetherefore studied the relationship between cardiac tro-ponin I (cTnI) and T (cTnT) and LV dysfunction in earlyseptic shock.  Methods:  Forty-six patients were consecutively en-rolled, fluid-resuscitated, and treated with cat-echolamines. Cardiac markers were measured at studyentry and after 24 and 48 h. LV function was assessed bytwo-dimensional transesophageal echocardiography.  Results:  Increased plasma concentrations of cTnI ( > 0.4  g/L) and cTnT ( > 0.1  g/L) were found in 50% and 36%,respectively, of the patients at one or more time points.cTnI and cTnT were significantly correlated ( r   0.847;  P  < 0.0001). Compared with cTnI-negative patients, cTnI-positive subjects were older, presented higher AcutePhysiology and Chronic Health Evaluation II scores atdiagnosis, and tended to have a worse survival rate anda more frequent history of arterial hypertension orprevious myocardial infarction. In contrast, the twogroups did not differ in type of infection or pathogen, orin dose and type of catecholamine administered. Con-tinuous electrocardiographic monitoring in all patientsand autopsy in 12 nonsurvivors did not disclose theoccurrence of acute ischemia during the first 48 h ofobservation. LV dysfunction was strongly associatedwith cTnI positivity (78% vs 9% in cTnI-negative pa-tients;  P   < 0.001). In multiple regression analysis, bothcTnI and cTnT were exclusively associated with LVdysfunction ( P  < 0.0001). Conclusions:  These findings suggest that in septicshock, clinically unrecognized myocardial cell injury isa marker of LV dysfunction. The latter condition tendsto occur more often in severely ill older patients withunderlying cardiovascular disease. Further studies areneeded to determine the extent to which myocardialdamage is a cause or a consequence of LV dysfunction. © 2000 American Association for Clinical Chemistry Inadequate myocardial performance, characterized by leftventricular (LV) 4 systolic depression and diastolic dilata-tion, is a common and early complication of septic shock (1 ,  2) . Studies in humans  (3 ,  4)  strongly argue against anischemic srcin of sepsis-induced cardiac injury. How-ever, a dysfunctional microcirculation that produces re-gional flow disturbances and abnormal tissue oxygen-ation is a hallmark of septic shock, which may causerelative ischemia in various organs, including the heart (5 ,  6) . Moreover, regional myocardial ischemia may well be present in septic patients with identifiable coronaryrisk factors or coexistent coronary artery disease.Cardiac troponins I (cTnI) and T (cTnT) are cardiospe-cific markers of prognostic value in acute myocardialinfarction (AMI)  (7–9) , unstable angina  (10 ,  11) , acutechest pain  (12–15)  myocarditis  (16) , cardiac trauma  (17) ,and perioperative cardiac complications  (18) . Recently,cTn positivity has been documented in patients with heartfailure of nonmyocardial ischemic srcin  (19 ,  20)  and in aheterogeneous population of critically ill patients in med-ical  (21 ,  22) , surgical  (23) , and pediatric  (24)  intensivecare units (ICUs). Increased cTn concentrations have beendescribed in the plasma of patients with sepsis  (22)  andseptic shock  (25 ,  26)  in association with an increasedmortality. Departments of   1 Clinical Chemistry,  2 Intensive Care, and  3 PathologicalAnatomy, Academic Hospital Vrije Universiteit Brussel (AZ-VUB), B-1090Brussels, Belgium.*Address correspondence to this author at: Academic Hospital VrijeUniversiteit Brussel (AZ-VUB), Department of Clinical Chemistry, Laarbeek-laan 101, B-1090 Brussels, Belgium. Fax 32-2-4775047; e-mail bdr@vub.ac.be.Received February 3, 2000; accepted February 28, 2000. 4 Nonstandard abbreviations: LV, left ventricular; cTn, cardiac troponin;AMI, acute myocardial infarction; ICU, intensive care unit; CK-MB, creatinekinase MB isoenzyme; CRP, C-reactive protein; TEE, two-dimensional trans-esophageal echocardiography; APACHE II, the Acute Physiology and ChronicHealth Evaluation II; ECG, electrocardiography; and  P c ,  P  value corrected forthe number of comparisons (Bonferroni adjustment). Clinical Chemistry  46:5650–657 (2000)  Enzymes and ProteinMarkers 650  The pathophysiological mechanism; the clinical, func-tional, and biochemical correlates; and the prognostic signif-icance of increased cTn concentrations in septic shock re-mainpoorlyunderstood.Therefore,acohortofpatientswithearly fully resuscitated septic shock was prospectively stud-ied with the following aims: ( a ) to determine the prevalenceof increased cTnI and cTnT in septic shock; ( b ) to comparecTn-positive and -negative patients in terms of demographicdata, clinical presentation, outcome, presence of other bio-chemical markers of cardiac injury [creatine kinase MBisoenzyme (CK-MB) mass determination; EC 2.7.3.2] orsepsis [procalcitonin and C-reactive protein (CRP)]  (27 ,  28) ,and anatomopathological findings in nonsurvivors; ( c ) tocorrelate cTn concentrations with in vivo LV dysfunction asassessed by two-dimensional transesophageal echocardiog-raphy (TEE). Materials and Methods patients The study was approved by the Committee for Ethics inHuman Research of the Academic Hospital of the VrijeUniversiteit Brussel. Informed consent was obtained fromthenextofkinofeachpatient.Forty-sixpatients(ages,18–93years; median, 66 years; male-to-female ratio, 30/16) inwhom septic shock was diagnosed within 4 h of clinicalonset were consecutively enrolled. Septic shock was definedaccording to consensus guidelines  (29)  as sepsis with hypo-tension unresponsive to fluid resuscitation and evidence of organ hypoperfusion or dysfunction, namely, ( a ) hypoten-sion, defined as systolic pressure   90 mmHg, or reducedfrom baseline by   40 mmHg; and ( b ) two or more of thefollowing criteria: temperature  38 °C or  36 °C; heart rate  90 beats/min; respiratory rate  30/min or hyperventila-tion with  P co 2  32 mmHg; white blood cell count  12 000/  L,  4000/  L, or  10% immature cells.Exclusion criteria included the presence of any cardio-thoracic event within 1 month before inclusion (coronaryinsufficiency, cardiothoracic trauma or surgery, cardio-pulmonary resuscitation, cardioversion, or endo-, myo-,or pericarditis) and immunosuppressed state [treatmentwith steroids, bone marrow or organ transplant recipi-ents, leukopenia (white blood cell count   1000/  L) orneutropenia (polymorphonuclear granulocyte count  500/  L), hematological malignancy, and AIDS], and amedical condition considered to be irreversible or lethalwithin 24 h after admission.All patients had indwelling radial (arterial line kit; Ar-gon) and balloon-tipped pulmonary artery catheters (Ed-wards Swan-Ganz model 97-120-7F; Baxter Healthcare) and,if required, were mechanically ventilated in volume- orpressure-controlled modes (Servo 900C ventilator; SiemensElema) under continuous analgesic sedation with midazo-lam and fentanyl. All patients received routine resuscitationtherapyforsepticshock,includingfluidadministrationwithcrystalloids and colloids, dobutamine to maintain cardiacindex   4 L    min  1   m  2 , and dopamine and/or norepi-nephrine to maintain mean arterial pressure   65 mmHg.After blood and various biological specimens were collectedfor microbiological analysis, all patients initially received broad-spectrumantibioticsconsistingofacombinationofanaminoglycoside with either a fourth-generation cephalospo-rinorciprofloxacin.Antibiotictreatmentwasadjustedbasedon culture results. clinical and functional investigations The Acute Physiology And Chronic Health Evaluation(APACHE) II score  (30)  was used to determine the initialseverity of illness. An inotrope score  (31)  was used to adjustfor relative catecholamine dependency. This score took intoaccount the type and dose of adrenergic agent(s) used andwas obtained for each patient after the desired endpoints forresuscitation were achieved. When a patient received acombination of adrenergic drugs, the inotrope score wascalculatedasthesumofthescoresforeachindividualagent.Any histories of cardiomyopathy, AMI, arterial hyperten-sion, diabetes mellitus, or peripheral arteriopathy were re-trieved from the patients’ medical files.After hemodynamic stabilization, a baseline 12-leadelectrocardiography (ECG) and a TEE were performed.Heart rate and rhythm were continuously monitored.ECG tracings were printed every 4 h and wheneverarrhythmias occurred. ECG findings matching myocar-dial ischemia were defined as flattened, inverted, orabnormally tall T-waves; a horizontal or sloping S-Tsegment depression; or a ST-segment elevation. LV dys-function was defined echocardiographically as the con-comitant presence of an increased LV end-diastolic diam-eter (  60 mm) and volume (  120 cm 3 ), the presence of regional and global LV hypokinesia, and a LV fractionalarea contraction of   0.4 under inotropic support  (32) . Thesame investigator (D.N.N.) performed all TEE measure-ments and assessed all ECG tracings while being blindedto the results for the cardiac markers. anatomopathological observations Survival was defined as leaving the hospital alive andable to resume all previous daily activities. Wheneverpossible, immediate postmortem examinations were per-formed in nonsurvivors to obtain myocardial tissue forlight microscopy examination. One fragment of LV freewall was fixed in 100 mL/L formalin, embedded inparaffin, and stained with hematoxylin-eosin, Massontrichrome, and Congo red. Sarcoplasmic fibrils and con-traction band necrosis were studied with a peroxidasetechnique (Dako Envision Systems) using a desmin anti- body (monoclonal anti-desmin II, 53-kDa desmin proteinspecificity; ICN Pharmaceuticals)  (33)  and with Heiden-hain’s iron hematoxylin, respectively  (33 ,  34) . All sampleswere assessed by the same pathologist (C.G.) blinded tothe results of the cardiac markers. blood sampling and biochemical assays Arterial blood samples were collected in lithium-heparinMonovettes (Sarstedt) on admission at the ICU and after Clinical Chemistry  46, No. 5, 2000  651  24 and 48 h. Samples were centrifuged immediately at3000  g  for 10 min (Hettich Zentrifugen), and the plasmawas aliquoted and stored at   70 °C. cTnI was measured by the Stratus II cTnI fluorometric enzyme immunoassay(Dade Behring)  (35 ,  36) , cTnT by the Elecsys 2010 Tropo-nin T STAT immunoassay (second-generation immuno-globulins; Roche), CK-MB mass by the Abbott AxSYMCK-MB microparticle enzyme immunoassay (Abbott Lab-oratories), procalcitonin by the LUMItest PCT immunolu-minometric assay (BRAHMS Diagnostica), and CRP bythe Vitros CRP slide enzymatic immunoassay (Johnson & Johnson Clinical Diagnostics). Cutoff values consideredindicative of cardiac injury when equaled or exceededwere 0.4   g/L for cTnI, 0.1   g/L for cTnT, and 9.0   g/Lfor CK-MB mass. Cutoff values for positivity were 0.5  g/L for procalcitonin, and 10 mg/L for CRP. statistical analysis Statistical tests were performed two-tailed using theGraphPad Prism, Ver. 2.0 (GraphPad Software). Differ-ences in prevalences and medians were determined by thetwo-tailed Fisher exact test and the Mann–Whitney  U  -test,respectively, and were considered significant at  P   0.05.In case of   k   comparisons, a corrected  P  value ( P c ) wascomputed by multiplying the  P  value by a factor  k  (Bonferroni adjustment). Differences in diagnostic infor-mation conferred by cTnI and cTnT status were assessedtwo-tailed by the McNemar exact test for paired observa-tions. Nonparametric correlations between two variableswere calculated by Spearman correlation. Multiple linearregression analysis (Windows 8.0; SPSS) assessed theability of cTnI and cTnT concentrations to independentlypredict LV dysfunction after adjustment for other vari-ables.  P  0.05 was considered statistically significant. Results causes and outcome of sepsis We prospectively followed 46 consecutively recruited pa-tients with septic shock. Pneumonia was the most frequentunderlying cause, followed by peritonitis and urosepsis(Table 1). Most infections (29 of 46, or 63%) were caused bygram-negative bacteria alone or in combination with otherpathogens (Table 1). In 39 patients (85%), one or morepathogenic infectious agents could be isolated or identified(Table 1). In the remaining seven patients, no microorgan-ismscouldbecultured,butinallofthemtheinfectiousfocuswas confirmed either clinically or at autopsy. Survivors (n  25) and nonsurvivors (n    21) did not differ in terms of underlying etiology of septic shock (Table 1) or sex ratio(Table 2). Nonsurvivors were older than survivors andtended (significance lost after Bonferroni correction) to havehigher APACHE II and inotrope scores. The two groups didnot differ in type or dose of specific catecholamines admin-istered or in a previous medical history of AMI or arterialhypertension (Table 2). cardiac markers in septic shock: clinical,functional, and biological correlates A high prevalence of increased cardiac markers (cTnI,cTnT, and CK-MB mass) was observed on admission atthe ICU and during the next 2 days. cTnT, cTnI, andCK-MB mass concentrations were above the cutoffs in50%, 36%, and 41% of the patients, respectively, at one ormore time points (Table 3). In marker-positive patients,the peak concentrations (median, interquartile range)were 1.4  g/L (0.8–6.8  g/L) for cTnI (n  23); 0.66  g/L(0.19–1.51   g/L) for cTnT (n    16); and 20.3   g/L(10.9–37.1   g/L) for CK-MB mass (n    19). A closecorrelation existed between cTnI and cTnT values ( r   0.847;  P   0.0001); a less strong but still significant corre-lation was found between cTnI and CK-MB mass ( r   0.618;  P   0.0001) and between cTnT and CK-MB massconcentrations ( r  0.525;  P  0.0001; Fig. 1). The temporalchanges of cTnI plasma concentrations did not differmeaningfully in survivors compared with nonsurvivingpatients (data not shown). Overall, the prevalence of increased cTn or CK-MB concentrations, if anything,tended to decrease with observation time (Table 3).cTnI-positive and -negative patients were next comparedin terms of demographic, clinical, and biological markers(Table 4). The male-to-female ratio did not differ accordingto cTnI status. cTnI-positive patients were on average older,had higher APACHE II scores on admission, and tended Table 1. Origin of septic shock and microbiological findings. Septic shockPatients with septic shock, nAll(n  46)Survivors(n  25)Nonsurvivors(n  21) Type of infectionPneumonia a  34 17 17Peritonitis b  6 4 2Urosepsis c  3 2 1Wound infection d  1 1Catheter sepsis c  1 1Meningitis c  1 1Type of pathogenGram-negative 22 13 9Gram-positive 6 4 2Mixed (  1 type of gram-negative)7 4 3Anaerobe 1 1Viral 2 1 1Yeast 1 1Not identified 7 2 5 a  The diagnosis of pneumonia was based on clinical and chest x-ray findings inall patients. Twenty patients grew pathogens on quantitative culturing of endo-bronchial secretions obtained by bronchoalveolar lavage or protected specimenbrush, 13 patients had similar microorganisms in bronchial aspirate and bloodcultures, and in 1 patient, pneumonia was confirmed at autopsy. b  Peritonitis was diagnosed preoperatively in all patients, two of them alsohaving positive blood cultures. c  Blood cultures were positive in patients with catheter sepsis, urosepsis, andmeningitis. d  The diagnosis of wound infection was made clinically. 652  ver Elst et al.: Cardiac Troponins in Septic Shock  (significancelostafterBonferronicorrection)tohaveaworseoutcome in terms of survival and to present more often witha history of AMI and arterial hypertension (Table 4), but notof diabetic arteriopathy or cardiomyopathy (data notshown). A significant correlation was found betweenAPACHE II score on admission and peak cTn concentra-tions ( P    0.0004 for cTnI, and  P    0.001 for cTnT bySpearman correlation). The inotrope score, dose of specificcatecholamines administered, and gram-negative srcin of shock were similar in both groups (Table 4). Regardless of cTnI status, atrial fibrillation, atrial flutter, and supraventric-ular tachycardia were frequently documented (data notshown). However, none of the patients presented signs of acuteischemiaonECGduringthefirst48hafterinclusioninthe study (Table 4). TEE disclosed a LV dysfunction in 18(78%) of cTnI-positive patients but in only 2 (9%) cTnI-negative patients ( P  0.0001; Table 4).All cTnI-negative patients were also cTnT negative, butseven subjects who were cTnI positive at some time pointremained cTnT negative ( P  0.03 by McNemar’s exact test).CK-MB mass did not occur more frequently in cTnI-positivecompared with cTnI-negative patients (Table 4). Markers of severity of septic shock, such as procalcitonin  (27)  and CRP (28) , were positive on admission in 91% and 98% of cases,respectively, and became positive in all patients after 24 h.Overall, cTnI-positive and -negative patients did not differin procalcitonin and CRP concentrations (Table 4). In survi-vors, procalcitonin concentrations after 24 h were similar tothose on admission but tended to decrease after 48 h ( P  0.01; Friedman-test). In nonsurvivors, procalcitonin andCRP values did not change significantly after 48 h. Clinicaland biochemical differences between cTnT-positive and-negative patients were similar to, albeit somewhat lesspronounced than, those found between cTnI-positive and-negative subjects (data not shown).After the distribution of maximal cTnI concentrationswas normalized by logarithmic transformation, multiplelinear regression analysis was used to assess the indepen-dent predictor ability of cTnI for LV dysfunction afteradjustment for the following independent variables: age,gender, survival, APACHE II and inotrope scores, dose of norepinephrine administered, and history of AMI and arte-rial hypertension. A highly significant association ( P  0.0001) between cTnI values and LV dysfunction on TEEwas found. After adjustment for LV dysfunction, no signif-icant correlation between cTnI concentrations and any of theothervariablestestedwasobservedevenafterthenumberof variableswasreducedtofive(LVdysfunction,age,survival,APACHE II score, norepinephrine dose). Similar resultswere found when cTnT concentrations were log-trans-formed and analyzed in the same way (data not shown). Table 2. Patient characteristics. CharacteristicsAll(n  46)Survivors(n  25)Nonsurvivors(n  21)  P  DemographicMale-to-female ratio 30/16 18/7 12/9 NS a,b  Age, median (interquartile range), years 66 (54–74) 54 (49–68) 72 (66–75) 0.001 c,d  ClinicalAPACHE II score, median (interquartile range) 24 (20–30) 21 (18–24) 27 (22–30) 0.018 c,e  LV dysfunction, n (%) 20 (43) 8 (32) 12 (57) NS b  Inotrope score, median (interquartile range) 4 (2–6) 3 (2–4) 6 (3–7) 0.032 c,e  Catecholamine dose, f   median(interquartile range),   g    kg  1   min  1 Dopamine 3 (0–10) 0 (0–5) 5 (0–10) NS c  Dobutamine 5 (0–10) 5 (0–5) 10 (0–15) NS c  Norepinephrine 0.1 (0.0–0.3) 0.1 (0.0–0.3) 0.2 (0.0–0.5) NS c  Medical history, n (%)AMI 10 (22) 4 (16) 6 (28) NS b  Arterial hypertension 13 (28) 5 (24) 8 (38) NS b  a  NS, not significant. b  Fisher’s exact test used for assessing differences between survivors and nonsurvivors. c  Mann–Whitney   U  -test used for assessing differences between survivors and nonsurvivors. d  P  c   0.01 after correction for the number of comparisons (n  10; Bonferroni adjustment). e  P  c , not significant after correction for the number of comparisons (n   10; Bonferroni adjustment). f   At the time of TEE. Table 3. Prevalence of cardiac markers. a  Time point cTnI b  cTnT c  CK-MB mass d  Admission 18/46 (39) 14/42 (33) 13/46 (28)24 h after inclusion 20/45 (44) 13/40 (32) 13/45 (29)48 h after inclusion 11/41 (27) 11/34 (32) 5/41 (12)  1 time point 23/46 (50) 16/45 (36) 19/46 (41)  2 time points 17/45 (38) 15/41 (37) 11/45 (24) a  Values given as n (%). b–d  Cutoff values for positivity:  b  cTnI  0.4   g/L;  c  cTnT  0.1   g/L;  d  CK-MBmass   9.0   g/L. Missing data attributable to insufficient serum sample or todeath of patient within 48 h after onset of septic shock. Clinical Chemistry  46, No. 5, 2000  653  anatomopathological findings In nonsurvivors, the elapsed time from study entry todeath averaged 8 days (range, 1–30 days). Multiorganfailure was the most common cause of death. Threepatients died of a late cardiac event (Table 5). Autopsieswere performed on seven cTnI-positive and five cTnI-negative subjects. Two of these patients died of a cardiacevent. Postmortem investigation revealed a LV free wallrupture in a cTnI-negative patient who underwent pro-longed cardiac massage and repeated electrical cardiover-sion after developing sudden ventricular fibrillation onday 3. Autopsy showed an extensive anterior myocardialinfarction in a cTnI-positive patient who died from intrac-table cardiogenic shock on day 7 after inclusion. Nopostmortem investigation could be performed on a cTnI-positive patient who died of a cardiac arrest on day 3without any sign of acute ischemia on ECG recordings.Microscopic examination of the heart excluded the pres-ence of myocardial infarction in 10 other patients (6 cTnIpositive and 4 cTnI negative) and documented the pres-ence of similar aspecific myocardial changes in marker-positive and -negative patients. These included limitedareas of elongated myocardial fibers, hypertrophied car-diomyocytes, and slight interstitial edema as well as smallclusters of “wavy” myocytes with condensed sarcoplasmand preserved nuclei. Myocardial infarction and inflam-matory cell infiltration could not be demonstrated. Con-traction band necrosis was seen in three of six cTnI-positive patients (50%) and one of four cTnI-negativepatients (25%;  P   0.05). Interstitial fibrosis was docu-mented in one cTnI-positive patient with a decreased LVfunction and in one cTnI-negative patient with normal LVfunction. One cTnI-positive patient with abnormal LVfunction had interstitial amyloidosis. Discussion This study confirms and extends earlier observations on thehigh prevalence of increased cTn concentrations in immu-nocompetent patients with acute resuscitated septic shock (23 ,  25 ,  26) . In addition, this study demonstrates a highlysignificant association in this clinical situation between thepresence of these specific and sensitive markers of myocar-dialcellinjury (7–24) andLVdysfunctionobjectifiedbyTEE.In our series of 46 patients, 50% displayed increased cTnIvalues at some time point during the first 48 h after admis-sion to the ICU, in agreement with 68% (18 of 26) cTnTpositivityreportedinaprevioussmallerstudybySpiesetal. (23)  on cardiac markers in early sepsis. In our hands, cTnIwas a more sensitive marker than cTnT or CK-MB; never-theless, the concentrations of these three cardiac markerswere significantly correlated, supporting release from dam-aged myocytes as an explanation for the increased plasmaconcentrations, rather than spuriously increased test results.The observed differences in marker prevalences may derivefrom differences in diagnostic performance between thevarious assays used or from differences in stability, struc-ture, myocyte content, and cellular release between thecardiac markers tested. The higher sensitivity of cTnI com-paredwithcTnTcontrastswithobservationsinotherclinicalsituations  (15) . Because different cTn isoforms are recog-nized to various degrees in certain assays, the possibilityshould be considered that different cTn fragments may bereleased in different pathophysiological circumstances  (37) .In agreement with previous observations in critically illpatients, cTn positivity was weakly associated with hos-pital mortality  (22 ,  23) . Moreover, the temporal evolutionof cTn concentrations did not contribute to a betterassessment of poor outcome  (22) . Clinically unrecognized Fig. 1. Correlation between cardiac markers. ( Top  ), cTnI and cTnT (n  116;  r   0.847;  P   0.0001); ( middle  ), CK-MB massand cTnI (n  132;  r   0.618;  P   0.0001); ( bottom  ), CK-MB mass and cTnI (n  116;  r     0.525;  P    0.0001; Spearman correlation). All samples were takenwithin the first 48 h after inclusion. 654  ver Elst et al.: Cardiac Troponins in Septic Shock
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks