Crosswords

The cost-effectiveness of competing strategies for the prevention of recurrent peptic ulcer hemorrhage

Description
The cost-effectiveness of competing strategies for the prevention of recurrent peptic ulcer hemorrhage
Categories
Published
of 10
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.
Related Documents
Share
Transcript
  The Cost-Effectiveness of Competing Strategies forthe Prevention of Recurrent Peptic UlcerHemorrhage Joshua Ofman, M.D., M.S.H.S., Joel Wallace, Pharm.D., M.P.H., M.B.A., Enkhe Badamgarav, M.D., M.P.H.,Chiun-Fang Chiou, Ph.D., James Henning, M.S., and Loren Laine, M.D.  Zynx Health, Subsidiary of Cedars-Sinai Health System, Beverly Hills, California; Departments of Medicineand Health Services Research, Cedars-Sinai Health System, Los Angeles, California; TAP PharmaceuticalProducts, Lake Forest, Illinois; and Division of Gastrointestinal and Liver Diseases, University of SouthernCalifornia, Los Angeles, California OBJECTIVE:  Several strategies exist for the prevention of recurrent ulcer-related hemorrhage, yet the cost-effective-ness has not been evaluated and remains uncertain. The aimof this study was to compare the cost-effectiveness of com-peting management strategies considering both nonsteroidalanti-inflammatory drugs status and the accuracy of   Helico-bacter pylori  (  H. pylori ) testing. METHODS:  Decision analysis was used to compare the cost-per-recurrent hemorrhage prevented for 11 strategies over 1yr. Clinical and costs estimates were derived from a sys-tematic review of the medical literature and the MedicareFee Schedule and Drug Topics Redbook. Sensitivity anal-yses were performed for important variables. RESULTS:  The test/retest eradication strategy with mainte-nance proton pump inhibitor therapy for  H. pylori -negativepatients was most effective (prevention of recurrence in96.0%). The test/retest eradication strategy with maintenancehistamine-2 receptor antagonist therapy for  H. pylori -negativepatientswasleastcostly($1070).Thetest/reteststrategiesweredominant with average cost-effectiveness ratios of $1118–1310/recurrent hemorrhage prevented with maintenance anti-secretory therapy. The average cost-effectiveness ratios for“selective”  H. pylori  eradication strategies with maintenanceantisecretorytherapywere$1263–1673.Themodelwasrobustto varying estimates over prespecified ranges. CONCLUSIONS:  Test/retest strategies for  H. pylori  are cost-effective for the prevention of recurrent ulcer-related hem-orrhage because they maximize  H. pylori  detection anderadication, resulting in fewer recurrent hemorrhages andfewer patients requiring antisecretory therapy. (Am J Gas-troenterol 2002;97:1941–1950. © 2002 by Am. Coll. of Gastroenterology) INTRODUCTION Upper GI hemorrhage (UGIH) represents a life-threateningmedical condition, with a mortality rate of approximately4–10% (1, 2). The most common cause of nonvaricealUGIH remains peptic ulcer disease, and the two majorcauses of peptic ulcer are nonsteroidal anti-inflammatorydrugs (NSAIDs) and  Helicobacter pylori  (  H. pylori ) infec-tion.  H. pylori  infection has been associated with 70–90%of peptic ulcers, and cure of   H. pylori  infection results in adecreased 1-yr ulcer recurrence rate from 70% to 2–20%(3–5).Although existing guidelines have focused on  H. pylori eradication for peptic ulcer disease, the adherence to theseguidelines remains less than optimal (6). Moreover, theadherence to accepted standards of practice in patients withUGIH is entirely unknown. Traditional management of ul-cer-related UGIH has relied upon treatment with mainte-nance antisecretory therapy despite a 19–30% rate of recur-rent ulcer and 9–10% rate of recurrent hemorrhage (7, 8).Recent data, however, provide evidence of the effectivenessof   H. pylori  eradication for preventing recurrent ulcer bleed-ing over 12–24 months (9–13). Recent pooled data analysesdemonstrate that  H. pylori  eradication may result in a 59–100% relative risk reduction for recurrent ulcer hemorrhagecompared with antisecretory therapy (14), and a 78–100%relative risk reduction compared with placebo (15).Although several competing strategies exist for the long-termmanagement of ulcer-related hemorrhage, the cost-effectivenessof competing strategies has not been evaluated to inform clinicalcare and resource allocation decision making. Moreover, the con-sideration of   H. pylori  and NSAID status and the accuracy of diagnostic testing as determinants of outcome have not beenevaluated. The objective of this study was to develop a decisionanalytic model to estimate the cost-effectiveness of competingmanagement strategies for ulcer-related hemorrhages consideringNSAID status and the accuracy of   H. pylori  testing. MATERIALS AND METHODS The population of interest is a hypothetical cohort of pa-tients with an initial ulcer-related hemorrhage, and the per- T HE  A MERICAN  J OURNAL OF  G ASTROENTEROLOGY  Vol. 97, No. 8, 2002© 2002 by Am. Coll. of Gastroenterology ISSN 0002-9270/02/$22.00Published by Elsevier Science Inc. PII S0002-9270(02)04261-2  spective of the economic analysis is that of a third-partypayer. This analysis does not consider patients with esoph-ageal variceal bleeding or stress-related mucosal hemor-rhage or patients with refractory or recurrent UGIH.A decision tree was developed incorporating base-caseestimates of the most likely clinical scenarios. The timehorizon of the analysis was 1 yr. The model incorporatedcost and probability estimates. All analyses were performedusing DATA decision analysis software (DATA, Version3.5, 2000, TreeAge Software, Boston, MA). Outcomes The outcomes evaluated included the average costs perpatient for each strategy and the effectiveness of each strat-egy as measured by the proportion of patients with recurrentulcer-related hemorrhage at 1 yr.  Decision Model  The decision analysis model evaluated 11 managementstrategies (Table 1). In these strategies, the patients pre-sented with initial ulcer-related hemorrhage, were hospital-ized, and underwent upper endoscopy. Three strategies in-corporated empirical treatments without  H. pylori  testing:  H. pylori  eradication, maintenance histamine-2 receptor an-tagonist (H2RA) treatment, and maintenance proton pumpinhibitor (PPI) treatment. The eight remaining strategieswere combinations of eradication and maintenance therapybased on the results of a biopsy urease test (Campylobacter-like organism) during the index endoscopy. Four of thesestrategies were  “ test-and-treat ”  strategies — testing for  H. pylori  infection and treating with eradication therapy forthose infected and maintenance antisecretory therapy withH2RAs or PPIs for those testing negative. Two of these fouremployed  “ selective ”  H. pylori  eradication in only thoseinfected patients not taking concomitant NSAIDs. Four ad-ditional strategies used a test/retest strategy under whichpatients with initial negative  H. pylori  tests underwent re-testing with a carbon-labeled urea breath test (CUBT). Weassumed that if patients were placed on antisecretory ther-apy before retesting, the antisecretory therapy would bediscontinued for 1 – 2 wk before the test to avoid false-negative tests. Similarly, two of these test/retest strategiesemployed  “ selective ”  H. pylori  eradication in only thoseinfected patients not taking concomitant NSAIDs. In allstrategies, patients receiving  H. pylori  eradication therapyreceived con fi rmatory CUBT 8 wk after completion of ther-apy.The physician was assumed to be unaware of false-pos-itive or false-negative test results; thus, Bayesean analysiswas built into the model to assess the accuracy of diagnostictests based on underlying  H. pylori  status (pretest preva-lence). Patients treated with  H. pylori  eradication therapyreceived a regimen of amoxicillin 1000 mg  b.i.d. , clarithro-mycin 500 mg  b.i.d. , and lansoprazole 30 mg  b.i.d.  for 14days followed by a con fi rmatory CUBT. Patients who con-tinued to test positive for  H. pylori  received a course of quadruple drug therapy (bismuth subsalicylate 525 mg q.i.d. , metronidazole 500 mg  q.i.d. , tetracycline 500 mg q.i.d. , and lansoprazole 30 mg  b.i.d. ) for 14 days. Patientswho tested  H. pylori  negative with CUBT or completedquadruple drug therapy were assumed by the physician to becured and received no further anti-  H. pylori  therapy.Patients treated with maintenance antisecretory therapyreceived either lansoprazole 30 mg daily for 6 wk followedby lansoprazole 15 mg daily for the remainder of 1 yr or  Table 1.  Treatment Strategies for Decision ModelEmpirical strategies Empirical eradication therapyEmpirical maintenance PPI therapyEmpirical maintenance H2RA therapyTest-and-treat strategies Eradication therapy for all  H. pylori -positive patients and maintenance PPI therapy for all  H. pylori -negative patientsEradication therapy for all  H. pylori -positive patients and maintenance H2RA therapy for all  H. pylori -negative patients “ Selective ”  eradication therapy for  H. pylori -positive patients not receiving NSAIDs andmaintenance PPI therapy for all  H. pylori -negative and NSAID patients. Patients switched toselective cyclooxygenase-2 inhibitors if chronic NSAIDs required “ Selective ”  eradication therapy for  H. pylori -positive patients not receiving NSAIDs andmaintenance H2RA therapy for all  H. pylori -negative and NSAID patients. Patients switchedto selective cyclooxygenase-2 inhibitors if chronic NSAIDs requiredTest/retest eradication strategies Patients with initial negative CLO get retested with CUBT. Eradication therapy for all  H. pylori -positive patients and maintenance PPI therapy for all  H. pylori -negative patientsPatients with initial negative CLO get retested with CUBT. Eradication therapy for all  H. pylori -positive patients and maintenance H2RA therapy for all  H. pylori -negative patientsPatients with initial negative CLO get retested with CUBT.  “ Selective ”  eradication therapy for  H. pylori -positive patients not receiving NSAIDs and maintenance PPI therapy for all  H. pylori -negative and NSAID patients. Patients switched to selective cyclooxygenase-2inhibitors if chronic NSAIDs requiredPatients with initial negative CLO get retested with CUBT.  “ Selective ”  eradication therapy for  H. pylori -positive patients not receiving NSAIDs and maintenance H2RA therapy for all  H. pylori -negative and NSAID patients. Patients switched to selective cyclooxygenase-2inhibitors if chronic NSAIDs required 1942 Ofman  et al.  AJG – Vol. 97, No. 8, 2002  ranitidine 300 mg daily for 6 wk followed by ranitidine 150mg daily for the remainder of 1 yr.In all treatment strategies, it was assumed that patientreport of NSAID usage was accurate based on patient his-tory, as there is no practical and reliable method for approx-imating  “ actual ”  NSAID use. Based on current guidelinesand practice patterns, it was assumed that physicians wouldattempt to discontinue NSAIDs in patients without the ne-cessity to remain on them, and patients having to remain onNSAIDs were switched to selective cyclooxygenase-2 in-hibitors.In all strategies, it was assumed that patients with arecurrent ulcer-related hemorrhage were hospitalized andthat the hemorrhage occurred approximately 6 months afterthe index bleed based on the mean time to recurrence in thepublished literature (7, 13, 16, 17). Despite the initial pre-vention strategy, all patients with a recurrent ulcer-relatedhemorrhage received maintenance PPI therapy for the re-mainder of the 1-yr follow-up period.All patients with an ulcer-related hemorrhage wereassumed to have a follow-up visit with a gastroenterolo-gist and two follow-up visits with their primary carephysician over 1 yr. Patients with a recurrent ulcer-related hemorrhage received an additional visit to a gas-troenterologist. It was assumed that ulcer-related hemor-rhages occurred with a ratio of 2:1 for duodenal ulcers togastric ulcers. Patients with gastric ulcers underwent fol-low-up endoscopy 6 – 8 wk after ulcer healing to assessfor underlying malignancy. Sensitivity Analysis One-way sensitivity analyses were performed to evaluatethe effects of varying costs and probability estimates overprespeci fi ed ranges. Two-way sensitivity analyses were per-formed on the most clinically signi fi cant or potentially in- fl uential variables. Clinical Inputs and Probability Estimates Estimates of   H. pylori  prevalence, NSAID status, eradi-cation rates, test characteristics, and rates of recurrentulcer-related hemorrhages were derived from a system-atic review of published literature in the MEDLINE andHEALTHSTAR computerized bibliographic databases(Table 2). We reviewed English-language articles from1985 to the present, and articles were identi fi ed by re-viewing selected bibliographies. When the literature hada range of probabilities, we used a weighted average toplace more emphasis on the results of studies with largersample sizes, and we chose estimates to bias the modelagainst the  H. pylori  eradication strategies. Where noevidence existed, a panel of expert gastroenterologistswas consulted, and estimates were tested over a widerange of uncertainty in sensitivity analysis.  Table 2.  Probability Estimates Used in the Decision ModelVariableBase-CaseEstimate(%)RangeTested(%) ReferencesPrevalence of   H. pylori  only 37 13 – 51 18 – 21Prevalence of   H. pylori  and NSAID 31 16 – 58 18 – 21Prevalence of NSAID only 14 7 – 25 18 – 21Probability of   H. pylori  eradication with triple therapy 85 45 – 98 9Probability of   H. pylori  eradication with quadruple therapy 80 45 – 98 9, 13, 16, 17, 22Probability of NSAID discontinuation 70 50 – 90 *Probability of recurrent hemorrhage in those testing  H. pylori  positive withouttreatment26 5 – 70 7, 8Probability of recurrent hemorrhage in those testing  H. pylori  positive after  H. pylori  eradication1 0 – 10 9Probability of recurrent hemorrhage in those testing  H. pylori  positive withmaintenance H2RA7 4 – 15 7 – 9, 12Probability of recurrent hemorrhage in those testing  H. pylori  positive withmaintenance PPI6 4 – 15 12, 23Probability of recurrent hemorrhage in those testing  H. pylori  negativewithout treatment26 5 – 70 13, 16, 17, 22,24Probability of recurrent hemorrhage in those testing  H. pylori  negative withmaintenance H2RA7 4 – 15 7 – 9, 12Probability of recurrent hemorrhage in those testing  H. pylori  negative withmaintenance PPI6 4 – 15 12, 23Sensitivity of rapid urease test (CLO) 76 45 – 93 25 – 32Sensitivity of urea breath test 95 90 – 99 28, 29, 33 – 36Sensitivity of serology 90 80 – 95 28 – 31, 37 – 47Speci fi city of rapid urease test (CLO) 93 80 – 99 25 – 32Speci fi city of urea breath test 95 80 – 99 28, 29, 33 – 36Speci fi city of serology 80 70 – 90 28 – 31, 37 – 47 * Probability and range assigned by expert panel. 1943 AJG – August, 2002  Competing Strategies for Peptic Ulcer Hemorrhage   Prevalence of   H. pylori  and NSAIDs Studies of patients with UGIH have found that 62 – 83% of patients with UGIH or ulcer-related hemorrhage were in-fected with  H. pylori , and 25 – 65% were taking NSAIDs(18 – 21). Patients both infected with  H. pylori  and usingNSAIDs made up 16 – 58% of all those with UGIH. Patientsinfected with  H. pylori  only made up 13 – 51%, whereasNSAIDs users alone made up 7 – 25% of patients experienc-ing an UGIH. Thus, we incorporated base-case estimates for  H. pylori  infection only in 37% (13 – 51%),  H. pylori  infec-tion and NSAID use in 31% (16 – 58%), NSAID use only in14% (7 – 25%), and neither NSAID use nor  H. pylori  infec-tion in the remaining 18% (0 – 58%). These ranges weretested in sensitivity analysis.  Effectiveness of Treatment In placebo-controlled trials of maintenance PPI and H2RAtherapy for the prevention of recurrent ulcer-related hemor-rhage, patients were not tested for  H. pylori  status (7, 8). Themixed  H. pylori -positive and -negative placebo groups hadrebleeding rates of 10 – 36%. Because of the paucity of published data, and to bias the model against  H. pylori eradication strategies, we assumed that the rates from themixed population observed in these trials would apply to the  H. pylori -positive and  H. pylori -negative populations. Thebase-case estimates for the rebleeding rates of both  H. pylori -positive and -negative patients treated with placebowere estimated to be 26%, with a range of 5 – 70% tested insensitivity analyses. Because the rebleeding rates in  H. pylori -positive and -negative patients may not necessarilybe interdependent, and there are scant published data re-garding this issue, we tested varying ranges independentlyin these groups in separate sensitivity analyses.The trials of H2RA treatment for the prevention of re-current ulcer-related hemorrhage included both placebo-controlled trials and head-to-head comparative trials ( e.g. ,H2RA  vs  PPI or eradication) (7 – 9, 12). The  H. pylori  statusof the patients in the placebo-controlled trial was unknown,but the rebleeding rates for patients were similar to  H. pylori- positive patients with rebleeding rates of 0 – 9% and4 – 8%, respectively. The base-case estimates for the re-bleeding rates of both  H. pylori -positive and -negative pa-tients treated with maintenance H2RAs were 7%, with arange of 4 – 15% tested in sensitivity analyses.Several head-to-head comparative trials have measuredthe effectiveness of PPI maintenance therapy for the pre-vention of recurrent ulcer-related hemorrhage comparedwith  H. pylori  eradication therapy. These trials includedonly  H. pylori -positive patients with 4% and 8% of patientsexperiencing a recurrent ulcer hemorrhage (12, 22). Becauseof the paucity of available data, it was assumed that  H. pylori -negative patients would have similar rebleedingrates. The base-case estimates for the rebleeding rates of both  H. pylori -positive and -negative patients on mainte-nance PPI therapy was estimated to be 6%, with a range of 4 – 15% tested in sensitivity analyses.The clinical trials of patients receiving  H. pylori  eradica-tion therapy for the prevention of recurrent ulcer-relatedhemorrhage were comparative trials with one trial also in-cluding a placebo group. The rebleeding rates for the pa-tients receiving  H. pylori  eradication therapy were 0 – 4% (9,13, 17, 18, 23). The base-case estimate was 1% based on theweighted average of the trials with a range of 0 – 10% testedin sensitivity analyses. Eradication rates from these studieswere 81 – 98%. The base-case was estimated to be 85% fortriple therapy and 80% for quadruple therapy, with a rangeof 45 – 98% tested in sensitivity analyses.  Diagnostic Testing In patients with UGIH, evidence suggests that the test char-acteristic of the CLO test is altered in the presence of bloodand is not as sensitive in patients with UGIH. Based on theresults of several studies of patients with UGIH, the base-case estimates for the sensitivity of the CLO test was 76%(45 – 93%) (24 – 32) with a speci fi city of 93% (80 – 99%) (28,29, 33 – 36). Cost Estimates Medication costs were obtained from the 2000 Drug TopicsRed Book of average wholesale prices. Costs for diagnosticprocedures and physician services were obtained from the2000 American Medical Association Current ProceduralTerminology codebook and the 2000 Medicare Reimburse-ment Fee Schedule (Table 3). Total costs were calculatedusing the sum of costs for procedures and services in eachtreatment arm. RESULTS The most effective strategies were the test/retest  H. pylori eradication strategies (Table 4). Testing patients with aninitial CLO test followed by retesting all  H. pylori -negativepatients with a CUBT, followed by  H. pylori  eradication inthose infected and maintenance PPI therapy in  H. pylori -negative patients prevented recurrent ulcer-related hemor-rhage in 96.0% of patients. The test/retest  H. pylori  eradi-cation strategy with maintenance H2RAs for  H. pylori -negative patients prevented recurrent ulcer-relatedhemorrhage in 95.7% of patients. Empirical antisecretorytherapy prevented recurrent ulcer-related hemorrhage in 93 – 94% of patients, and empirical  H. pylori  eradication therapyprevented recurrent ulcer-related hemorrhage in 90% of patients.The three least costly strategies included two test/retesteradication strategies and one test-and-treat strategy (Table4). The test/retest  H. pylori  eradication strategy with main-tenance H2RAs for  H. pylori -negative patients was the leastcostly strategy with an average cost of $1070 per patient.The similar strategy without retesting for initial  H. pylori -negative patients resulted in a marginal cost of $19.The results of the cost-effectiveness analysis suggest thattesting patients with an initial CLO test followed by retest- 1944 Ofman  et al.  AJG – Vol. 97, No. 8, 2002  ing all  H. pylori -negative patients with a breath test, fol-lowed by treating  H. pylori -positive patients with eradica-tion therapy and  H. pylori -negative patients with eithermaintenance H2RA or PPI therapy were the dominant strat-egies (more effective and less costly). The cost-effective-ness ratios (cost/recurrent ulcer-related hemorrhage pre-vented) for the two test/retest strategies including eithermaintenance H2RA or PPI therapies were $1,118 and$1,310, respectively. The marginal cost-effectiveness be-tween the test/retest strategy using maintenance H2RA andthe test/retest strategy using maintenance PPI was $64,931per recurrent ulcer-related hemorrhage prevented.Among the test/retest strategies, those employing eradi-cation in all  H. pylori -infected patients (regardless of NSAIDs use) were generally less costly and more effectivethan those employing  “ selective ”  H. pylori  eradication inonly infected patients not taking NSAIDs. Similarly, in thetest-and-treat strategies, those employing  H. pylori  eradica-tion in all infected patients had lower average cost-effec-tiveness ratios than those strategies employing selective  H. pylori  eradication in NSAID-negative patients. Even whenwe assumed that the  “ selective ”  strategies only incurred thecosts of the CLO test in those not using NSAIDs, they didnot become the preferred strategy. Sensitivity Analysis We performed sensitivity analyses to evaluate the effect of varying cost and probability estimates on the cost-effective-ness of competing strategies (Table 5). The comparisonreference strategy of initial CLO, followed by retesting  H. pylori -negative patients with urea breath test, then treating  H. pylori -positive patients with eradication therapy andtreating  H. pylori -negative patients with maintenance H2RAtherapy was used for the analysis. The model was sensitiveto a few critical variables. If the CUBT costs more than $144(base-case $104), then retesting  H. pylori -negative patientswith CUBT is no longer the optimal strategy. The  H. pylori eradication rate for triple therapy must decrease to below60% before the test/retest strategy loses its cost-effective-ness advantage over the test-and-treat strategy.Regarding effectiveness, we assumed that the recurrenthemorrhage rate in  H. pylori -negative patients without treat-ment was 26%. This rate would need to be less than 18% forempirical  H. pylori  eradication to become most cost-effec-tive, and more than 48% for test-and-treat therapy to bemost cost-effective. We assumed that successful  H. pylori eradication in those infected reduced the recurrent ulcer-related hemorrhage rate to 1%. If the recurrence rate isgreater than 4%, then the test-and-treat strategies are morecost-effective than the test/retest strategies. If the recurrencerate is greater than 7%, then the empirical strategies aremore cost-effective than the test/retest strategies. Similarly,if the prevalence of   H. pylori  in ulcer-related hemorrhagefalls below 60%, then the test/retest strategy loses its cost-effective advantage compared with the test-and-treat strat-  Table 3.  Cost Estimates Used in the Decision ModelProcedure or Treatment CPT/DRG Code Cost ($) Range Tested ($)CLO test (cost of upper endoscopy with biopsy — cost of endoscopy) 63 10 – 95Upper endoscopy with biopsy 43239 302Upper endoscopy 43234 239Triple therapy (Prevpac) AWP 252 125 – 500Quadruple therapy AWP 123 50 – 250Maintenance lansoprazole (1 yr) AWP 1339 750 – 1450 † Maintenance lansoprazole (6 mo) AWP 673 300 – 725 † Maintenance ranitidine (1 yr) AWP 609 550 – 986 ‡ Maintenance ranitidine (6 mo) AWP 339 300 – 494 ‡ GI follow-up visit 99243 142 75 – 300Primary care follow-up visit 99213 58 25 – 100Hospital care for UGIH* 4959 2500 – 7500GI hemorrhage 174 3964Endoscopy, upper GI hemorrhage 43255 357Initial hospital care 99223 184Subsequent hospital care (4 days) 99233 375Hospital discharge 99238 79Urea breath test 104 45 – 200Breath test 83013 93Drug administration and sample collection 83014 11Serology test 86677 8.03 8 – 25Outpatient GI visit with endoscopy and biopsy with CLO 975 450 – 1500Upper endoscopy with biopsy 43293 302Of  fi ce consultation 99245 254Facility fee 419 AWP  average wholesale price; CPT  current procedural terminology; DRG  diagnosis-related group.* Assumed DRG, which includes any setting within the hospital, including intensive care unit stays, as we recognize that a proportion of patients with stable ulcer-relatedhemorrhage may be admitted to the intensive care unit. †  Lansoprazole 30 mg for 1 yr  $1365 and 6 mo  $684. ‡  Ranitidine 300 mg for 1 yr  $986 and 6 mo  $494. 1945 AJG – August, 2002  Competing Strategies for Peptic Ulcer Hemorrhage
Search
Similar documents
View more...
Tags
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