Originally Published MDDI October 2005
Several methods of leak detection can be used to ensure the safety of medical equipment. It's important to understand the properties of each to choose the right method or combination of methods.
By Claes Nylander
Medical devices like the Berlin Heart axial pump may need new leak-test methods.
Methods of leak detection used by medical device manufacturers are evolving to meet increasingly stringent quality standards. Testing for leak integrity is critical in the device industry because product failure presents significant hazards to patients and healthcare professionals and liabilities for manufacturers.
Medical devices often involve barriers between fluids or gases. These barriers must function effectively to protect the patient and, quite often, the medical device itself. Some devices serve as pathways that deliver or extract fluids to or from a precise location in the body. If a leak exists along the pathway, fluids could be accidentally delivered to the wrong location, or the wrong fluids could be extracted, inciting dangerous consequences.
The most common methods of leak detection include pressurized air and tracer gas systems, and there is increasing focus on using hydrogen as a tracer gas. Hydrogen has gained wide acceptance in leak detection in the aeronautics, automobile, and HVAC industries. It also possesses several distinct attributes that make it suitable for a number of medical applications.
Hydrogen-based leak detection has shown high sensitivity to leaks and is becoming more popular for medical devices.
Leak detection in the medical equipment industry often needs to address special requirements:
• Because of critical safety issues, in some cases leak testing must be performed on 100% of the product, without being destructive or corrosive to the parts. Sample testing is insufficient.
• Frequently, medical products contain materials that are flexible and elastic,
Small devices, like this Incor axial ganz, pose a particular challenge at the port and tubing sections. Those areas may have leaks that are difficult to detect.
allowing fluctuations in volume that can distort the results of indirect leak detection methods that measure changes in pressure in the object's interior.
• Because sequential testing is often required on separate but contiguous components, residual buildup of clinging tracer gases can cause problems during subsequent tests.
Dry-Air and Wet Methods
A number of leak-detection methods are used in manufacturing medical consumables, medical implants, and medical instruments. The most common methods are the various dry-air test methods. In these tests, an object is pressurized (or evacuated), then monitored to detect any leakage. Leaking is indicated by a change in pressure.
The dry-air technique acquired its name because it replaced the wet methods that are still prevalent in other industries. Wet testing is an inexpensive and straightforward approach. It involves elevating the air pressure inside an object and then submerging it into a water bath or brushing soap bubbles on the exterior and watching for the formation of bubbles. Such techniques are unacceptable for most medical applications because the wet testing medium can contaminate the device. Further, the process is highly susceptible to operator error and is difficult to automate.
Pressure Decay. The original dry-air method is pressure decay, in which the test part is pressurized and then isolated from the pressure source. Because air moves from a high-pressure area to a low-pressure area, any decrease in pressure indicates the presence of a leak. Algorithms then convert pressure changes into an approximation of the rate of leakage.
Pressure-decay testing can have a long cycle time, which might be problematic for high-volume production facilities. Because of the compression heating effect, the initial pressure measurement must be delayed until the temperature has stabilized. After stabilization, an additional time interval must elapse between this initial measurement and a second reading. This interval allows sufficient time for a measurable amount of leakage to occur. The longer the interval, the more likely small leaks will be detected. However, the relationship between test duration and sensitivity is not linear. For example, if a hole size were reduced by half, the testing would take at least four times longer to detect it. Therefore, time delay presents a practical limit to the sensitivity of pressure-decay testing.
A similar method, vacuum decay, evacuates air from the test object. For this method, it is the loss of vacuum that indicates a leak, rather than the loss of pressure. Unlike pressure decay, the vacuum-decay method does not require a delay for temperature stabilization because there is no compressive heating effect. However, it does take more time to evacuate the air from an object than to fill it with high-pressure air, although the difference is minimal for the small volumes of most medical products.
To address the long cycle time, a faster approach has been developed, known as the differential-pressure method. With this method, a leak-free reference volume is pressurized along with the test part. A transducer then reads the difference in pressure between the nonleaking reference and the test item over time. Algorithms convert this difference into a volumetric measure of the leakage rate.
Mass Flow. Another dry-air method, the mass-flow method, monitors air flowing into a pressurized object to compensate for any loss of air due to a leak. This flow can come either from a reference volume reservoir that is pressurized along with the test part or from an air-supply line controlled by a regulator. A sensitive flowmeter measures how much air is required to maintain the pressure. This volume is theoretically equivalent to the amount of leakage.
Mass flow is considerably faster than the various pressure-decay methods, as only a single measurement is required. The sensitivity does not depend as much on testing volume as it does for the decay methods. Mass flow is used to test tubing and other medical components, because it can simultaneously test for occlusions in an open-ended test part.
Effectiveness of Dry-Air Methods. The various dry-air methods have gained wide acceptance in the medical equipment industry. These methods can be easily automated, and testing data can be acquired to analyze quality assurance and to implement process improvement. All of the methods utilizing pressurized air and vacuum work relatively well for objects that are small and have minimal interior volume. The dry-air methods are economical because the equipment itself is relatively inexpensive and air does not need to be purchased; compressed air or a vacuum is generated on-site.
However, the flexible and elastic materials used in many medical devices and consumables present a challenge for dry-air testing methods. Any expansion or contraction of the object causes changes in the internal air pressure and distorts the test results. Manufacturers may attempt to control this effect by physically restraining the object to minimize expansion during pressurized leak testing. For example, balloon catheters, collection bags, blood bags, IV bags, and nourishment bags are often sandwiched between two plates to minimize expansion. Unfortunately, these plates cover a significant percentage of the bag's exterior surface, potentially blocking actual leaks that are temporarily in contact with the plates.
Recently, more-sensitive methods using tracer gases such as helium and hydrogen have proven to be effective for testing the leak integrity of medical equipment. Tracer gases are necessary because pressure-decay systems do not have sufficient sensitivity to detect extremely fine leaks in components that require a bacteria barrier. Mass flow and pressure decay can detect leaks as slow as 0.005 cm3/sec and 0.0001 cm3/sec, respectively. But the most sensitive tracer gas method—helium mass spectrometry in a vacuum chamber—can detect leaks as slow as 1 × 10–12 cm3/sec.
There are several reasons why tracer-gas methods are able to detect much smaller or slower leaks:
• The testing equipment is highly sensitive and can detect minute levels of the gas.
• Tracer-gas methods detect the actual gas that passes through leaks, rather than measuring the derivative effects of the pressure and flow changes caused by leaks.
• Derivative effects such as pressure decay depend not only on the amount of air left in the test object, but also on temperature and volume. A tracer-gas test result is a direct measurement of what comes out. Hence, it is independent of temperature and volume.
For example, say you have a leak in a bucket. You can measure the leak by counting droplets or by measuring the water level inside the bucket. Obviously the first method is more sensitive.
The need for high leak-detection sensitivity in certain medical equipment applications is the principal reason that manufacturers choose a tracer-gas method. Like the dry-air methods, the tracer-gas methods can also be automated and connected to information systems aimed for higher productivity and quality. Unlike the dry-air methods, the tracer-gas methods are not affected by the compression heating caused by pressurized air, so there is no need to delay the test procedure while pressure and temperature are stabilized.
Moreover, the tracer-gas methods are unaffected by the elasticity or plasticity of the test object and, therefore, the test object does not need to be restrained.
Helium Mass Spectrometry. The most common tracer-gas method in the medical equipment industry is a helium-based test method. It involves pressurizing the test part with helium or a helium-air mixture within a vacuum chamber. The chamber surrounding the part is then evacuated, creating a vacuum that invites the helium to pass through any leaks into the chamber. A mass spectrometer, specifically tuned to detect helium, samples the vacuum chamber and ionizes any helium present. Even trace amounts of the helium are readily detected.
The helium vacuum chamber method provides the highest sensitivity of any method used in medical equipment manufacturing. However, the apparatus is complex and cost-prohibitive. In most cases, it is overengineered for all but the most demanding applications. A mass spectrometer is an expensive and delicate instrument that requires specialized training and regular maintenance. Also, the vacuum chamber must be well-constructed to maintain a tight seal, and vacuum pumps must be purchased and maintained.
Helium vacuum chamber testing, like the dry methods, is an integrity test that detects the total leakage of an object. It does not determine the location of a leak, or even the number of leaks in the object. For applications where leak location is important (e.g., repairs and process improvement), a sniffer can be employed instead of a vacuum chamber. The sniffer is a tube that uses a mechanical vacuum pump to draw air to the mass spectrometer. The sniffer scans the exterior of a helium-filled object to directly detect any leakage points. The sniffing approach does not require the construction and maintenance of a vacuum chamber, but it is 10,000 times less sensitive than the vacuum chamber method.
An advantage of the helium method is that the tracer gas is injected with minimal pressure, so there is no risk of deforming or destroying bags or other flexible materials. Moreover, using a sniffer can pinpoint the location of leaks, unlike dry-air pressure testing that can only measure the total leakage of an entire object.
Microelectronic Hydrogen Tests. In response to the high cost and complexities of helium mass spectrometry, a relatively new technology, hydrogen leak testing, has emerged as a viable alternative that provides high sensitivity, but is more cost-effective. This technology uses a microelectronic sensor system, which can detect leaks that are more than 100 times smaller than those that can be detected using pressure-decay systems.
As with the helium-based method, there are two fundamental methodologies for hydrogen-based testing. First, a highly sensitive integrity test can be performed on an object as a whole by placing it inside an accumulation chamber and filling the object with hydrogen tracer gas. Because hydrogen dissipates quickly and the background concentration in air is low, the accumulation chamber operates at atmospheric pressure, unlike the vacuum-chamber system required for helium mass spectrometry.
Second, a hydrogen probe can pinpoint the location of leaks by manually or robotically scanning the exterior of an object filled with hydrogen tracer gas, in a manner similar to the helium sniffer procedure. The hydrogen probe, however, is not a sniffer per se, because no mechanical sniffer pump is required to draw gases through a tube to a remote testing instrument. The hydrogen probe is therefore less maintenance-intensive. Because the sensor is at the tip of the probe, the hydrogen detector responds quickly to the presence of elevated levels of the element. The robust probe requires no service, and if the probe becomes damaged, the operator can easily replace it in a few minutes.
The hydrogen sensor employs a semiconductor design similar to a field-effect transistor. Hydrogen molecules adhere to the surface of the sensor, where they dissociate into hydrogen ions (protons). These protons then penetrate the lattice structure of the sensor, where they cause a change in an electrical field and trigger a signal that passes through a cable to a signal processor in the programmable testing device. The sensor responds selectively only to hydrogen.
It is worthwhile to point out that the microelectronic sensor system is significantly less expensive than a helium mass spectrometer. Once other costs are considered, such as maintenance and the price of helium versus hydrogen, the difference is palpable.
Properties of Hydrogen
Hydrogen has a number of attributes that make it well suited to testing for leaks in medical equipment (see Table I). Hydrogen spreads easily throughout a test object and penetrates leaks readily. It also vents away significantly faster than any other tracer gas does, thereby minimizing the chances of background buildup. Because hydrogen does not stick to surfaces, it is useful for sequential tests (e.g., the testing of multiple lumens in a catheter or multiple compartments of an implantable medical device). The penetration of tracer gas through the polymer material of medical objects may be larger than the flow through a very small leak. This seems to set a limit to how small a leak you can find. However, penetration through the polymer itself takes time, so if the test is done quickly no gas will have time to penetrate the material and emerge on the outside. The flow of gas through even a very small pore is established very quickly and emerges at the outside almost immediately.
Table I. Properties of hydrogen, helium, and air.
Hydrogen leak detection employs a readily available mixture of 5% hydrogen (H2) and 95% nitrogen (N2). The mixture is certified as nonflammable to ISO 10156 and presents no safety risks. Primarily, the mixture is deployed as a shielding gas for welding, soldering, and brazing. Industrial gas suppliers usually provide 200-bar (2.900-psi) and 300-bar (4.350-psi) pressurized gas cylinders.
Hydrogen-based testing also meets the requirements of ISO 14001. Because hydrogen is a naturally occurring gas that is considered nontoxic, it has no adverse effects on the environment and can be produced anywhere with simple methods, thereby minimizing transportation costs and pollution.
Hydrogen is the least expensive of all tracer gases. And because of the increased interest in hydrogen fuel cells and related technology, the price of hydrogen is likely to steadily decrease. By contrast, helium is a scarce natural resource, which has seen price increases of about 10% each year recently.
Case Study: Testing an Implantable Device
The axial-flow pump Berlin Heart Incor, developed by Berlin Heart AG , presents a case study for deciding which leak detection method to use. This implantable blood pump has flexible components, making it ill suited for any test methods that involve significant overpressure. Joint areas, such as those at the tubing ports, may develop tiny leaks that can be difficult to detect. Furthermore, small devices often need low-viscosity testing. Low viscosity reduces the possibility that residual gases will trigger false-positives for leaks. On such sensitive equipment, a false positive could lead to a significant unnecessary expense.
For these reasons, the tracer-gas method using hydrogen may be the best option for the pump. Hydrogen is effective at testing the integrity of seals where tubing is connected to a port. A probe can scan these locations and pinpoint the exact source of leaks. Additionally, the elastic nature of the tubing could affect the test outcome.
Hydrogen can be easily flushed from the device after tests are completed, and its molecular viscosity detects small leaks or punctures that may not be detected with other methods.
Marrying Hydrogen and Dry-Air Pressure Methods
By providing high sensitivity and the ability to test elastic objects, the hydrogen method overcomes many of the shortcomings of dry-air pressure methods. Many manufacturers have begun to take advantage of the benefits of hydrogen testing to enhance or augment the testing performed by dry-air methods, such as pressure-decay testing.
Hydrogen and dry-air methods are compatible approaches that can be deployed interchangeably or in conjunction with each other, depending on the requirements of both the manufacturing process and quality standards.
In fact, the testing process, the test apparatus, the training requirements, and the cost of the hydrogen method closely match the characteristics of dry-air pressure methods, and therefore, manufacturers can easily incorporate hydrogen into their existing dry-air leak-testing systems or replace dry-air systems with hydrogen systems. For example, pressure decay and hydrogen are increasingly used in tandem: the pressure-decay system tests all parts to ensure leak integrity, and a hydrogen test is performed only on those parts that fail, pinpointing the number and location of leaks for quality assurance and process improvement.
Within the medical industry, product testing is an indispensable component of the device manufacturing process. In particular, leak testing has become vital to medical equipment manufacturers as quality standards have become increasingly stringent. Choosing the proper leak-detection method is an important and often complicated challenge. Manufacturers should carefully explore the various leak-testing methods on the market. There are many opportunities to improve quality assurance, lower testing costs, and minimize the risk of product failures.
Copyright ©2005 Medical Device & Diagnostic Industry