Instructions on installing the model:

Extract all files first. To be able to run the model all of the following files should be saved in the same folder:
AVRSim012001.exe (= the interface of the model)
CE-SAV05022007.dat (= the latest version of the CE-SAV input file)
CE-Pericardial.dat (= the latest version of the CE-Pericardial input file)
Allograft01062001.dat (= the latest version of the allograft input file)
AVR BPMP250607.dat (= the latest version of a new mechanical and bioprosthesis input file) 
NL.female.txt (=Dutch lifetable for females)
NL.male.txt (=Dutch lifetable for males)
US.female.txt (= US lifetable for females)
US.male.txt (= US lifetable for males)
UK.female.txt (= UK lifetable for females)
UK.male.txt (= UK lifetable for males)
BC.female.txt (= lifetable for females in British Columbia, Canada)
BC.male.txt (= lifetable for males in British Columbia, Canada)

Put all files in the same folder, otherwise the model will not function properly!  

The model currently consists of:
·         an interface (Figure 6) that allows the user to select a patient profile (age and gender; CABG is still inactive) and the simulation settings (valve types to simulate; number of patients to simulate e.g. 10,000; whether or not to take operative mortality into account);
·         an input file (Figure 7) that contains the evidence-based estimates of the occurrence of valve-related events and operative mortality for the different valve types, the background mortality (general Dutch population or US population) plus excess mortality estimates for patients after AVR.  

How does AVRSim work?
To open the model, double-click on AVRSim012001.exe, and the interface will open (Figure 6). Then, click on the INPUT button and select the CE-SAV05022007.dat – file or the CE-Pericardial.dat - file. Now you can view the input file (Figure 7). Note that it is possible to change the estimates of the occurrence of valve-related events and operative mortality in the input file by double clicking on the estimates you want to change, and simply typing in the estimate you prefer. Please check the changes you made by closing the input file (press the OK button), reopening it (press the INPUT button on the interface) and checking whether the change is still in effect. Furthermore, it is possible to change the background mortality by choosing one of the given lifetables.
To run the model, press the OK button of the input file. Then, enter on the interface the desired age, gender, number of simulations (e.g. 10,000) and valve type(s), and press the button START SIMULATIONS. It takes approximately 1-10 seconds for the model to finish the simulations, depending on your computer. Total life expectancy, event free and reoperation free life expectancy estimates plus standard errors appear on the interface. If you like more details on the output, press the VIEW OUTPUT button. This allows you to take a look at cumulative survival, freedom from event and reoperation, event-free and reoperation-free survival, a count of all valve-related events and their associated mortality, operative mortality, non-valve-related mortality etc.  

Small exercises to get you started:
Take for instance as input the CESAV input file (CE-SAV05022007.dat). Confirm the input parameters are equal to the parameters given in the example in the manual, and that the background is based on the US lifetables for males (ASRmales.txt). For these exercises, let’s simulate 65-year old male patients, say 10,000 times.

Keep in mind that the AVRsim program is written European style: this means figures behind comma’s describe decimals, and dots indicate thousands.  

Exercise 1.      Test the effect of higher (e.g. doubling) operative mortality on total life-expectancy.

Exercise 2.      Change background mortality or/and excess mortality, and study the effect? 

Exercise 3a.   If the risk of thrombo-embolism (TE) doubles, what does this mean for total life-expectancy, and for event-free life-expectancy? Does this change the reoperation-free life-expectancy?
              b.      If the risk of prosthetic valve endocarditis (PVE) doubles, what is the effect on total life-expectancy and event-free life-expectancy? Does this change reoperation-free life-expectancy (RFLE)? What happens when the reoperation rate changes? Does this affect RFLE?
             c.      Suppose mortality for TE increases to 80%. Which outcome parameter will be affected most?  

Answer 1.      First run the model using the CESAV input file (CE-SAV05022007.dat). Fill in: age ‘65’, Sex click ‘Male’, under ‘simulation settings’ click ‘bioprosthesis’, and simulate 10,000 patients. The button ‘include 1st operation’ should be switched ‘on’, in order to include the operative mortality of the primary operation. Click ‘START SIMULATIONS’. The life-expectancy (LE), event-free life-expectancy (EFLE), and (re)operation-free life-expectancy (RFLE) are respectively: 10.6; 9.2 and 9.8 years. 
         Now change the basic operative mortality in the input file: first open the input file, double-click on the operative mortality for bioprostheses (0.027), and change it to 0.054. Close the input file and reopen it to make sure that your change is still in effect. Then, run the model again. When the basic operative mortality is doubled (with increasing OR for age left unchanged), the LE, EFLE and RFLE are 10.0; 8.7 and 9.3. As could be expected, increasing the operative mortality negatively affects long-term outcome on all strata.
Please undo your changes in the input file in order to get the correct answers to the next exercises.

Answer 2.      In the input file, the background mortality can be changed by browsing to another file instead of the US lifetables. One can select one of the other files, such as ‘Mrmal90.cbs’ which is the dutch lifetable for males. Also other life tables can be incorporated in the model, after they have been downloaded and have been put in the same folder as the simulation program.
         To change the excess mortality, simply double-click on the figures for the different age-groups, and change them. For a 65-year old male, the baseline estimate in the present model with the CESAV input file is 1.2. Let’s increase this estimate by 25%, so replace 1.2 by 1.5. Now again fill in: age ‘65’, Sex click ‘Male’, under ‘simulation settings’ click ‘bioprosthesis’, and simulate 10,000 patients. Increasing this baseline estimate by 25% has quite some influence on the outcome parameters: LE, EFLE and RFLE are now respectively 9.6; 8.6 and 9.1.
Please undo your changes in the input file in order to get the correct answers to the next exercises. 

Answer 3a.     Please see the input file. Click the ‘thrombo-embolism’ subform. For thrombo-embolism, the linearized occurrence rate based on the CE-SAV dataset was 1.12 (which is equal to 0.0112 per patient year). Now double this value by changing this number into 0.0224, simulate 10,000 65-year old males again and view the output: LE 10.2; EFLE 8.5; RFLE 9.5. Obviously, the EFLE is most affected, far more than RFLE, and also more than LE. However, increasing the occurrence of an event that has no consequences for the number of reoperations, like TE, will have some effect on RFLE, just by reducing total LE.
Please undo your changes in the input file in order to get the correct answers to the next exercises.

             b.     Increasing the occurrence of an event with a relatively high frequency (such as thrombo-embolism), will have more effect on life-expectancy, than increasing the occurrence of a low-frequency event (such as PVE). Therefore, when the occurrence of PVE is doubled, the LE, EFLE and RFLE are respectively 10.5; 9.0 and 9.7. To check this, please go to the input file, click the ‘endocarditis’ subform, double the occurrence from 0.0033 into 0.0066, and simulate 10,000 65-year old men again. Even when the reoperation-rate would be doubled, this hardly has any effect on the RFLE. (LE, EFLE and RFLE are 10.5; 9.0 and 9.6, when both the occurrence of PVE and the reoperation-rate of PVE are doubled).
Please undo your changes in the input file in order to get the correct answers to the last exercise.

             c.     Please go to the input file, select the ‘thrombo-embolism’ subform again and change the mortality of the event from 0.419 into 0.8. Start the simulations, and view the output parameters. Life-expectancy (LE) is most affected: LE is 10.2, EFLE is 9.2 and RFLE is 9.5. 

Saving changes / creating a new input file
To save any changes you made in the input file, please click ‘Save Input’ in the input file. To name your own file, replace the asterisk (*) by the name you’d like to give to your input file, making sure the filename ends with ‘.dat’ and is saved in the same folder.      

Still to do...
·     Concerning the input
Like we state in this manual, microsimulation still has several limitations. Some of those limitations will be attacked in the near future. For instance, a model to incorporate additional CABG is under development, and we are in the process of adding more variables, such as renal failure, left ventricular function, cardiac rhythm, COPD etc. Up till now, the model is not capable of producing distributions of the input parameters, but work in this direction is in progress. Furthermore, new databases with high quality data on outcome after AVR are needed to create new input files for different valve types/brands, and for the more exact calculation of certain hazards of valve-related events, such as bleeding and endocarditis.
·         Concerning the output
We are in the process of changing the interface of the model, and would like to add a user-friendly output file that contains all information the clinician is interested in. Therefore, we would like to invite you to share your comments and suggestions with us in order to improve both the model and this website. This can be done by e-mail to This e-mail address is being protected from spam bots, you need JavaScript enabled to view it .

Thank you for your efforts!  

References
1. van Geldorp, M.W., et al. Usefullness of microsimulation to translate valve performance into patient outcome: patient prognosis after aortic valve replacement with the Carpentier-Edwards supra-annular valve. J Thorac Cardiovasc Surg, 2007;134:702-9.
2. Puvimanasinghe JP, Steyerberg EW, Takkenberg JJ, Eijkemans MJ, van Herwerden LA, Bogers AJ, Habbema JD. Prognosis after aortic valve replacement with a bioprosthesis: predictions based on meta-analysis and microsimulation. Circulation 2001;103:1535-41.
3. van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E. Assessment of a bleeding risk index in two cohorts of patients treated with oral anticoagulants. Thromb Haemost 1996;76:12-6.
4. http://www.cdc.gov/.
5. Steyerberg EW, Kallewaard M, van der Graaf Y, van Herwerden LA, Habbema JD. Decision analyses for prophylactic replacement of the Bjork-Shiley convexo-concave heart valve: an evaluation of assumptions and estimates. Med Decis Making 2000;20:20-32.
6. Puvimanasinghe JP, Takkenberg JJ, Edwards MB, Eijkemans MJ, Steyerberg EW, Van Herwerden LA, Taylor KM, Grunkemeier GL, Habbema JD, Bogers AJ. Comparison of outcomes after aortic valve replacement with a mechanical valve or a bioprosthesis using microsimulation. Heart 2004;90:1172-8.
7. Takkenberg JJ, Eijkemans MJ, van Herwerden LA, Steyerberg EW, Lane MM, Elkins RC, Habbema JD, Bogers AJ. Prognosis after aortic root replacement with cryopreserved allografts in adults. Ann Thorac Surg 2003;75:1482-9.
8. Takkenberg JJ, Eijkemans MJ, van Herwerden LA, Steyerberg EW, Grunkemeier GL, Habbema JD, Bogers AJ. Estimated event-free life expectancy after autograft aortic root replacement in adults. Ann Thorac Surg 2001;71:S344-8.
9. Takkenberg JJM, Puvimanasinghe J.P.A, Van Herwerden, L.A. et al. Prognosis after aortic valve replacement with SJM bileaflet prostheses: impact on outcome of varying thrombo-embolic hazard European Heart Journal 2001;3;Q27-Q32.