A



Dual Port, Single Lumen Peripheral IV Catheter

Kelly Baron

Erik Frazier

Riley Smith

April 29, 2005

Table of Contents

Specific Aims Page 3

Significance Page 4

Relevant Experience Page 7

Methods Page 8

Conclusions Page 17

Future Work Page 18

References Page 19

Appendix A Page 20

Appendix B Page 21

Appendix C Page 24

A. Specific Aims

Two types of peripheral midline IV catheters are available on the market today. The first is a dual port, dual lumen design that is used to deliver and collect fluids through a long tube that, after insertion, extends through the vein from the elbow to the shoulder. Because these devices are large and made of multiple plastic materials, they tend to be more costly and more complicated to insert. Therefore, only trained IV teams and experienced nurses are skilled enough to insert them into a patient. The more preferred choice of peripheral midlines in hospitals today is a single port, single lumen midline. This type of IV catheter is used much more frequently because it is cheaper and less complicated to insert. There are many limitations associated with these devices in that because they are much less complex than the dual port, dual lumen designs, they can be removed and reinserted much more frequently. Frequent needle sticking and insertion of the IV catheter may result in minor complications such as thrombosis, infection, edema and most commonly pain and discomfort for the patient. Therefore, the new peripheral IV catheter design will aim to cost-efficiently eliminate many of the complications involved with current single port, single lumen peripheral IV catheters on the market. The dual port device redesign will allow for delivery and collection of fluids through the same catheter lumen. Although the same risks for thrombosis and infection will still be present with the new device, the ultimate goal of the new design will help to decrease patient discomfort by decreasing the number of times the patient will need to be stuck with a needle. The new device will be very similar to current single lumen peripheral midlines on the market today but we will convert to a dual port design.

The goals of this Phase I program were to:

Specific Aim 1: Design a dual port, single lumen peripheral IV catheter similar to today’s common single port, single lumen design. The new design 1) is cost-efficient 2) just as easy to insert as current single port designs and 3) is biocompatible.

Specific Aim 2: Fabricate the dual port, single lumen IV catheter prototype using stereolithography and manual fabrication in sufficient quantity to support Phase I in-vitro testing.

Specific Aim 3: Test the Phase I prototype in vitro to assess functionality and basic biocompatibility.

Specific Aim 4: Evaluate the Phase I prototype in vivo during future acute animal experiments.

It is important to note that the dual port design will not be appropriate for use in situations where it is unacceptable to stop IV fluids, such as the OR, ER or ICU. Only patients with acute symptoms in need of fluids or blood draws will benefit from the new design.

B. Significance

The average yearly United States market for infusion intravenous (IV) devices exceeds 1.37 billion dollars and includes a combination of products produced in the United States and those sold overseas. Over the past five years alone, the market has grown 10.2 percent (Business Communications). On the current IV market, there are two main types of IV catheters: central lines and peripheral midlines. These devices are used to deliver essential IV medications and nutritional fluids and/or draw blood through either the subclavian vein or a peripheral vein in the bend of the elbow, respectively (Higa).

On average, ninety percent of all patients admitted to hospitals for care are in need of some type of IV therapy (Higa). Typically, central lines are used for patients who are in need of a long-term line, months or years, or who require short-term drug or antibiotic fluids that peripheral veins cannot withstand. Because central lines need to be inserted into the subclavian vein by a surgeon, they are very costly and have all of the inherent risks of any surgical procedure where a patient needs to be anesthetized. Midlines are generally used for the short-term administration of IV drugs and antibiotics and blood draws, but they can also be utilized for long-term use. They are used more frequently in the hospitals because they are much less costly than their central line counterparts and there are fewer risks involved with insertion because they are inserted into the peripheral veins of the arms. However, because peripheral veins are much smaller in diameter than the subclavian vein, one limitation to these devices is that thrombosis occurs much more frequently. Blood clots within the blood vessels are caused by injury to the blood vessel walls that result in platelets adhering to the vessel walls. The aggregation of platelets makes it impossible to draw blood and/or administer fluids through these veins. Some indirect causes of blood clots in midlines are if the catheter is too large for the vessel or is not secured properly, multiple venipuncture attempts, administration of medications incompatible with solution or of a high pH or tonicity, ineffective filtration, kinks in the IV tubing or if the venipuncture is carried out by an unskilled professional (IV Complications). Although there are not many risks involved with blood clots in the midlines, once a clot occurs, fluids can no longer be delivered or drawn through them. Therefore, the line must be removed and inserted again through another peripheral vein.

Administering fluids and drawing blood through these devices can be complicated by thrombosis and infection as described above. These complications may make the occurrence of reinsertion very high and replacing peripheral catheters can be a challenge if the patient’s venous access is poor or if reinsertion becomes a frequent task. Currently, patient discomfort experienced from frequent insertion of IV catheter midlines and sticks from blood draws during hospital stays are one of the most common complaints on venipuncture Press Ganey Scores (Better or Worse?). There have been consistent declines in patient ratings when patients were asked the questions “How well was the IV started?” and “How well was blood taken?” (Better or Worse?). In addition to the barriers of limited dexterity and sensitive touch created by today’s surgical gloves and the fewer number of IV insertions done by specialists, a single lumen IV catheter seems to be cause of increased patient discomfort because cost and simplicity cause them to be used most frequently. Single lumen IV catheters can only be used to either deliver fluids or draw blood, and never simultaneously through the same line. Therefore, patients in need of both IV fluids and blood draws must be stuck in both arms. Because the IV catheter is designed to last three to five days, most only need to be inserted once during that period if no complications or blood clots arise. On the other hand, most hospitals do not insert these single lumen IV catheters for daily blood draws and will instead stick the patient’s arm every time blood is drawn. Hypothetically, if a patient needs IV fluids and blood must be drawn twice a day during a four day hospital stay, the patient will be stuck around eight times for the blood draws plus those associated with the administration of IV fluids (one to two times under the previous assumption of no complications). Therefore, upon discharge on the fifth day, the patient’s arms will have been stuck around ten times during their visit. The insertion site may show irritation from skin preparation, activity of catheter site, skin contamination, catheter material, and from the chemicals/medications administered. When the vein/artery is irritated, the risk of catheter-related infection increases as well in addition to discomfort from bruises and inflammation. The new device, if used properly, could bring an eighty percent decrease to these needle sticks: two sticks verses ten, hypothetically.

There is an increasing need for an easier and more cost effective way to deliver fluids to and draw blood from patients through the same peripheral midline. It was essential to perform research directly on ways to incorporate both of these aspects of patient care into one dual port, single lumen IV catheter device. Although dual lumen midlines currently on the market eliminate most patient discomfort, cost and complications cause them to rarely be used. Instead, single lumen peripheral midlines are used to either deliver fluids or draw blood and although they are less costly and easier to use, the consequences of extreme patient discomfort made it vital for a new dual port, single lumen design to be developed that is cost-efficient and just as easy to use as current single port, single lumen peripheral catheters on the market.

C. Relevant Experience

The principal investigators for the proposed research project will be Kelly Baron, Erik Frazier, and Riley Smith. Sandra Gartner and associates at Vorp Laboratories will assist with design and development of the device, as experts in the field.

Principal Investigators: All three investigators are fourth year Bioengineering students at the University of Pittsburgh. Kelly Baron’s specific concentration is in the area of biomechanics. In 2003, she worked at the Musculoskeletal Research Center, under the advising of Dr. Savio Woo, Ph.D., where she helped to improve the accuracy of the laboratory’s laser micrometer system in determining the cross-sectional area of soft tissues. In 2004, she worked in the Neuromuscular Research Laboratory where she developed a program used to calculate the forces at the elbow and shoulder joints during baseball pitching. In addition to biomechanics, Baron has had classroom experience in the areas of biothermodynamics, biological signals and systems, ergonomics, biotransport, bioinstrumentation, and human physiology.

Erik Frazier has tailored his undergraduate bioengineering studies to the areas of biotechnology and artificial organs. Since this past summer he has been working at an internship at the Musculoskeletal Research Center under the advising of Dr. Savio Woo, Ph.D. There he developed a new laser scanning system to better determine the cross-sectional area and shape of soft tissues with concavities. Frazier’s classroom experience includes study in the areas of human physiology, biochemistry, biothermodynamics, biological signals and systems, biotransport, bioinstrumentation, and physiological dynamic systems.

Riley Smith focuses mainly on the area of BioMechanics within BioEngineering. In 2003 he worked with Dr. Debski on researching the variable responses to rotator cuff tears in different individuals. In addition to BioMechanics Smith has studied the areas of human physiology, biochemistry, biothermodynamics, biological signals and systems, biotransport, bioinstrumentation, and physiological dynamic systems.

D. Methods, Conclusions and Future Work

Specific Aim 1: Design a dual port, single lumen peripheral IV catheter similar to today’s common single port, single lumen design. The new design 1) is cost-efficient 2) just as easy to insert as current single port designs and 3) is biocompatible.

Our original design plan was to fabricate a dual port, dual lumen IV catheter that allows for simultaneous fluid delivery and fluid collection (Figure 1). However, the complexity involved in the device’s entrance into the vein would require that a sheath be used to introduce the catheter. Because this complexity would require a specialist for insertion, a dual port, single lumen catheter design was proposed.

[pic]

Figure 1: Original dual port, dual lumen design.

The new design will be easier to insert than dual port, dual lumen IV catheters currently on the market. It will be inserted in place of the widely used single lumen IV catheters, but will have a slightly longer hub portion. Two standard luer lock ports will be exposed outside the skin during use, one for fluid collection, and one for fluid delivery. Fluid delivery must be stopped and the line flushed before fluid is drawn, making the new device unsuitable for situations where interruption of drip is unacceptable. A standard luer lock design provides compatibility with all hospital luer-lock components in use today such as standard syringes and IV lines.

The insertion process is modeled after the insertion of single port, single lumen IV catheters so that it is possible for lesser-trained individuals to insert the new device. Current dual lumen IV catheters require highly trained individuals for insertion. Insertion of our dual port, single lumen IV catheter will not require highly trained individuals because advancing the device through the vein will no longer be needed. Our goal is to make our design simple enough that it would enable any hospital health care technician to perform an insertion. By doing so, we would not only reduce the cost down to that of single lumen catheters but we would enable a more economical distribution of the hospital’s resources and employees. Specialized individuals would not be paid for their time and could their experience elsewhere.

We do not expect any significant biocompatibility issues with our redesign because the materials have been used in the same manner for years. In addition to biocompatibility issues, we plan to use already tested materials for two other reasons. First, when packaged properly, they have been proven to effectively maintain sterility during their shelf life. The need for sterility is obviously important to prevent infections. Second, they have proven economical for one time use because of their disposability. Being disposable is one of our main design criteria because no money conscious hospital would want to invest in IV catheters that would need to be sterilized between uses, especially when so many disposable ones are currently available on the market.

Specific Aim 2: Fabricate the prototype dual port, single lumen IV catheter using stereolithography and manual fabrication in sufficient quantity to support Phase I in-vitro testing.

A prototype model was constructed in Solid-Works and it provided us with many valuable insights. We were able to make simple adjustments very easily, without investing any money to do so. We were able to get realistic information about our device (Do the pieces really fit together?). Once the design conformed to our specifications in Solid-Works, we were confidently able to finalize our design, order materials, and advance to the building of a prototype.

The modified and original design files were taken to The Swanson Center located in Benedum Hall and an initial rigid polymer SI 20 hub model was produced. SI 20 is around ten percent more rigid than the high-density polyurethane material used in standard catheters on the market (3D Systems). The SI 20 material is acceptable for prototype testing, but will be changed to polyurethane for future human testing or human use. Prototype development throughout the course of the project can be seen in Figure 1 of Appendix A with the final design represented as Design 3.

Fifty one and one quarter inch and fifty one and three quarter inch Medex® Optiva® 18G catheters were ordered. The one and one quarter inch catheter lumens were extracted from the Optiva® catheters and manually fabricated between two halves of the distal end of the device with epoxy. The two luer lock ports exactly modeled those that are currently used on the market to assure that the new device would be compatible with other components used in the hospital setting (i.e. luer-lock connections). Because the length of the catheter hub was extended slightly, the one and three quarter inch guide needles were used for insertion.

Specific Aim 3: Test the Phase I prototype in vitro to evaluate functionality and basic biocompatibility.

A Simulaids, Inc. IV training arm was used to test the functionality of our prototype. The arm is made of supple, resilient plastic with a human feel and the internal veins are analogized with latex tubing. The “veins” were connected to two bags filled with a blood solution so that the blood could flow through the veins. The full bag was raised above the arm, the other one set below and gravity was used to drive the blood flow (Figure 1 of Appendix B). We introduced our IV catheter with the needle designed for the Optiva® one and three quarter inch single port catheter and the device lumen was inserted one and one quarter inch internally. Once the vein was located, the prototype was inserted in exactly the same manner as the single port IV catheters widely used today. We ran six trials, each consisting of prototype insertion, fluid delivery, and blood draw (Figure 2 of Appendix B). For comparison, the Optiva® catheter was also inserted for six trials.

Three main successes can be attributed to the testing with the training arm. First, our prototype was just as easily inserted as current single port IV catheters on the market. Second, the prototype was capable of delivering IV fluids through both ports. Finally, the prototype was capable of drawing the blood analog into a syringe (Figure 3 of Appendix B). The limitations experienced from our prototype include the following. The prototype fabrication was not completely air tight, allowing some air bubbles in the blood draw samples. We had no prior IV experience which led to damage of the prototype during repetitive testing. The blood analog developed some aggregations in it, changing the flow properties.

Our prototype was also tested to ensure pressure performance at specific flow rates. The device was connected to a Harvard Apparatus Co. Compact Infusion Pump (Model 975), an Edwards Pressure Transducer and a SpaceLabs Pressure Monitor (Model 90603A) (Figure 4 of Appendix A). Using Equation 3, flow rates were determined at various infusion speeds by observing the time needed to pump 5 mL of water through the device (Table 1 of Appendix C).

5 mL / (t * 60 sec/min) = FR [3]

Where: t is the infusion time in seconds and FR is the flow rate

Peak pressure was measured for three trials at each flow rate for the: straight port, curved port and standard Optiva® 18G 1 ¼” catheter. Prototype peak pressure was averaged among trials (Tables 2, 3 and 4 of Appendix C), plotted against flow rate and compared to that of the standard (Figure 2).

[pic]

Figure 2: Pressure verses flow curve for the straight port, curved port and standard Optiva® 18G catheter.

It can be seen in Figure 2, as flow rate is increased variability between the prototype and the standard catheter increases. However, at smaller flow rates, the two devices perform much more similarly. Because the projected maximum flow rate of the targeted patient population is around 125 mL/hr or 2.08 mL/min, results show that there is no difference between our device in both the straight and curved portions and a standard catheter in the applicable range; our device is capable of performing at a standard peak pressure.

Given the nature of this design, it was also important to determine the amount of contamination that would occur in a blood sample. In order to quantify the contamination, two sample draw protocols were used. In both protocol a dye solution was run through the device from the curved port before a sample is taken, and the samples were quantified by spectral density. The first step of the process was preparing 10 mL of 5% Coomassie Blue (a dye commonly used to stain proteins) solution, and then diluting the solution, sequentially, 16 times with an equal volume of water. Three wells of a plate were filled with each of the solutions and the plate was then inserted into a BioRad Microplate Reader (Model 680) operating at a wavelength of 570 nm. The average optical density for each sample was used to determine a calibration curve relating optical density and concentration of solutions (Table 5 of Appendix C).

In order to draw the simulated blood samples, 2 mL of the darkest solution that the reader could measure was injected into a waste beaker through the curved port (Figure 5 of Appendix B). In the first protocol, the straight line was flushed with 2 mL of water while the second protocol was not flushed. A 5 mL sample of water was then drawn from a beaker for each protocol. These samples were then put into a microplate (again filling three wells) and the optical densities were found (Figure 6 of Appendix B).

From the optical density data gathered from the microplate reader, the greatest readable concentration was 4.8 x 10-3 %. This value became the maximum concentration in the calibration curve, but it was decided to use the second highest concentration (2.4 x 10-3 % solution) as the contaminant in the trials. The calibration curve, which is visible in Figure 3, shows a clearly linear relationship between concentration and optical density over the range of readable concentrations. The relationship between the two values is given by equation 1.

OD= 36610*[CB]-0.0161 [1]

Where: OD is optical density and [CB] is the concentration of Coomasie Blue

[pic]

Figure 3: Average Optical Density for each known concentration level (%). This functions as a calibration curve for Coomassie Blue. A linear regression of the curve indicates that OD=36610*[CB]-0.0161. The R2 value of the equation is 0.9962.

Equation 1 can be rearranged to solve for the unknown concentrations of the two protocol samples based on the average known optical density, as seen in equation 2

[CB]= (OD+0.0161)/36610 [2]

Inputting the average optical density for each protocol (0.0297 for protocol 1 and 0.0805 for protocol 2) into the equation, the concentration of Coomasie Blue in the samples were calculated to be 1.25 x 10-4 % and 2.63 x 10-4 %, respectively. These two concentrations, when divided by the concentration of the original contaminant give the percent contamination: about 5% for protocol 1 and 11% for protocol 2.

The high correlation for the optical density-concentration equation gives validity to the concentrations obtained using this method. The assessment is also strengthened by the use of 3 readings for each concentration and sample as the standard deviation of each optical density is much less than the related average. Additionally, the contamination seen in protocol 1 is considered to be insignificant.

Specific Aim 4: Evaluate the Phase I prototype in vivo during future acute animal experiments.

In order to ensure accuracy of our modeling and our assumptions regarding biocompatibility, we would follow our in vitro testing with in vivo testing on animals in the future. Our animal model would be a dog or a goat due to their relatively large size and common use in testing. The large size of the veins in the animals, which are comparable to human veins, would provide a much better test of the ease of use and likelihood of venipuncture than the in vivo model that we developed. Additionally, in using a living subject, the contents of a blood sample collected during administration of fluids could be compared to blood taken from the opposite appendage in order to validate in vitro result of having blood samples unaffected by the foreign fluids. Finally, by using an animal we would be able to detect any unforeseen biocompatibility issues without risking human health. A population of several (5-10) animals would have the device inserted into one of the front legs where it would remain for several days while glucose would be administered through the IV for the periods of time leading up to and including when blood is drawn. Blood would be extracted from each of the fore limbs from each subject and the glucose content from each limb is compared for every animal. This difference will tell us how much the use of the device effects blood tests results. Additionally, since the animal is under observation for several days we will be able to determine if the length of time that the catheter is in place affects the accuracy of the test results. If there is a time when the tests are no longer deemed to be accurate then the recommended life of the product would be altered to reflect that time. We will also be able to observe the insertion point on the animals to ensure that there is no inflammation and no infection from improperly sterilized devices.

Conclusions:

The ease of insertion and ability to deliver and collect fluids through the same lumen really makes this a viable product. The biggest difference in our product is its ability to draw blood and deliver fluids without reinsertion or disconnecting the IV and this was also proved possible. These three major successes really define the functionality of our product.

The limitations experienced during functionality testing can all be attributed to problems that would not exist in a medical setting. The fabrication of the prototype was not completely air tight, allowing air bubbles to collect in the blood draw samples. We had no prior IV experience, and the prototypes (as well as the final product) are designed for single use only so the combination of these two facts, led to damage of the prototype during testing. Finally, the blood analog developed some aggregations in it probably due to improper mixing. This analog would not be present in the medical field and further testing would have to be done to see if real blood would clot in our prototype.

The pressure/flow results reveal that the prototype is capable of delivering fluids at a maximum necessary flow rate. However, some error may exist causing a slight difference between our device and the standard, but this error can be contributed to error in the stereolithography prototyping process (rough surfaces, changing inner diameters, etc.).

The high correlation for the optical density-concentration equation gives validity to the concentrations obtained using this method. They show that as long as the line is flushed before blood is drawn, any contamination present will be negligible. Additionally, if the IV being administered does not contain any of the substances being tested for in a certain blood test, that it would be feasible to continue the IV drip while blood is drawn.

All test results reveal that our device specifications and specific aims were met throughout the course of this project. The device is capable of fluid delivery and collection through the same lumen and if used properly, will decrease patient discomfort typically experienced from frequent blood draws.

Future Work:

Small modifications could be implemented to the design of the device to decrease the difference in pressure drop between the straight and curved ports. By moving the luer lock of the curved port toward the patient and decreasing the severity of the curve in the path the associated resistance would be decreased, thus decreasing the necessary pressure to generate a desired flow (Figure 2 of Appendix A).

In order to move the device into production, the remainder of the parts must be designed to eliminate the usage of competitors’ products. This step includes the design of needle and flashback chamber to assist in the induction of the device, as well as designing a silicon lumen and retainer ring that will be compatible with our designed hub. Additionally, polyurethane models would need to be produced in order to determine the flow properties of the device when it is made out of the proper material. By testing a prototype made of polyurethane we can determine whether the differences in pressure-flow relationships between our device and the device already on the market were due to the roughness of the SI-20. Our fully redesigned device could then be used in animal trials to evaluate the in vivo usability of the device as described above.

References

(in order cited)

US Catheter Sales to Exceed $23 Billion by 2009, , December 11, 2004.

Higa, Lisa S, Infection Control Today, , December 11, 2004.

IV Complications, 3M Skin Health Program, < professionals/skinhealth/pdfs/iv_complications.pdf>, December 12, 2004.

Better or Worse? Patient’s Perceptions of Hospital Services, , December 11, 2004.

3D Systems, , April 15, 2005.

OPTIVA IV Catheter with OCRILON Polyurethane. Johnson & Johnson Gateway,

, December 11, 2004.

Appendix A

[pic]

Figure 1: Prototype design development.

[pic]

Figure 2: SolidWorks model of future design.

Appendix B

Figure 1: Testing set up shown with the full bag of blood raised above the arm. The latex tubing can be seen coming out of the right side of the arm. Our prototype has been inserted into a vein and has a syringe attached to one port and IV tubing attached to the other.

Figure 2: Close-up photo showing the inserted prototype again with a syringe and IV tubing attached.

Figure 3: Successful blood draw. The small air pocket at the top of the blood draw volume is due to the prototype not being air tight.

[pic]

Figure 4: Prototype Pressure/Flow Testing Setup

Figure 5: Syringes connected to the catheter in order to determine

contamination of sample. Image shows catheter prior to line being flushed.

Figure 6: Optical density microplate, each concentration of Coomassie Blue

fills three wells. Samples are on the bottom left wells.

Appendix C

|Gear |Flow Rate (mL/min) |

|1 |41.763 |

|3 |10.949 |

|5 |9.934 |

|7 |5.505 |

Table 1: Calculated flow rates for each respective infusion pump gear setting used.

|Pressure | | | | | |

|(Straight Port) | | | | | |

|Gear |P1 |P2 |P3 |Average |St. Dev. |

|1 |28 |27 |28 |27.667 |0.577 |

|3 |13 |12 |11 |12.000 |1.000 |

|5 |8 |8 |6 |7.333 |1.155 |

|7 |5 |6 |4 |5.000 |1.000 |

Table 2: Peak pressure, average pressure and standard deviation for straight port.

|Pressure | | | | | |

|(Curved Port) | | | | | |

|Gear |P1 |P2 |P3 |Average |St. Dev. |

|1 |35 |39 |42 |38.667 |3.512 |

|3 |11 |16 |16 |14.333 |2.887 |

|5 |10 |8 |7 |8.333 |1.528 |

|7 |5 |4 |4 |4.333 |0.577 |

Table 3: Peak pressure, average pressure and standard deviation for curved port.

|Pressure (Standard SP) | | | | | |

|Gear |P1 |P2 |P3 |Average |St. Dev. |

|1 |20 |20 |24 |21.333 |2.309 |

|3 |7 |9 |10 |8.667 |1.528 |

|5 |4 |8 |6 |6.000 |2.000 |

|7 |1 |5 |4 |3.333 |2.082 |

Table 4: Peak pressure, average pressure and standard deviation for standard catheter.

|  |Concentration (%) |

  |0 |7.63E-07 |1.53E-06 |3.05E-06 |6.10E-06 |1.22E-05 |2.44E-05 |4.88E-05 | |Trial 1 |0.025 |0.046 |0.048 |0.066 |0.163 |0.621 |0.85 |1.921 | |Trial 2 |0.024 |0.038 |0.045 |0.068 |0.167 |0.371 |0.79 |1.734 | |Trial 3 |0.027 |0.038 |0.045 |0.065 |0.165 |0.375 |0.814 |1.75 | |Average |0.025333 |0.04066 |0.046 |0.06633 |0.165 |0.455667 |0.818 |1.80167 | |S. D. |0.001527 |0.00461 |0.001732 |0.00152 |0.002 |0.143197 |0.030199 |0.1036 | |

Table 5: Optical density data is presented at each known concentration for all three trials. Average Optical density and Standard deviations have also been calculated for every density.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download