7. How are new devices tested for accuracy?
- 7.1 Quality requirements for the alternatives to the Hg manometers
- 7.2 Technical Verification
- 7.3 Clinical validation
- 7.4 Discussion
- 7.5 Recommendations
7.1 Quality requirements for the alternatives to the Hg manometers
In December 2007 the standard ISO 81060-1 "Non-invasive sphygmomanometers – Part 1: Requirements and test methods for non-automated measurement type” was published. This standard addresses all kinds of sphygmomanometers, "which, by means of inflatable cuffs, are used for the non-invasive blood pressure measurement by operator observation” (ISO 2007). Automated sphygmomanometers are addressed in a different standard issued in 2009: IEC 80601-2-30 "Medical Electrical Equipment – Part 2-30: Particular requirements for the basic safety and essential performance of automated non-invasive sphygmomanometers” (ISO 2009). The standard ISO 81060-2 "Non-invasive sphygmomanometers – Part2: Clinical validation of the automated measurement type (ISO in preparation). All three standards are expected to become European harmonized standards in the near future.
The ISO 81060-1 addresses requirements for the alternative non-automated sphygmomanometers. Because these requirements are identical for all possible manometers, they include requirements for accuracy of the cuff pressure measurement and for the resistance to vibration and shock. Some requirements are related to the specific needs of aneroid manometers. The ISO/CEN standards are non-mandatory but may be used as tools for checking the reliability of the alternatives to Hg sphygmomanometers and comply with the essential requirements of the medical device directive (93/42/EEC).
7.2 Technical Verification
Regular metrological testing is needed to ensure the accuracy of the blood pressure devices. Periodic maintenance and accuracy testing may be initiated by the manufacturers instructions or by legal measures (Germany, Austria, Czech Republic, and Slovakia). Statistical data on the percentage of failure of such verification exist only from ten and more years ago, at that time the number was between eight and ten percent per year (PTB-Mitteilungen, 1990). There is no indication that this number has dramatically changed.
The key element of the verification is the testing of the accuracy of the static pressure measurement by the manometer of the sphygmomanometer. In pressure steps of not more than 50 mmHg over the whole measuring range the error of the pressure measurement has to be determined. For this test a periodically calibrated reference manometer has to be used, usually a digital manometer utilising a piezo-resistant transducer. Mercury manometers are not appropriate for use as reference manometers because their resolution is not good enough and it is not easy to identify the meniscus of the mercury column in order to read exact values (less than 1.0 mmHg).
7.3 Clinical validation
Independent device accuracy assessment within a clinical setting is recommended before introduction and routine clinical use. Various protocols have been published to assess automated devices against a mercury sphygmomanometer during clinical use and these are referred to as clinical validation protocols. The International Protocol of the European Society of Hypertension (O'Brien et al. 2002) and the protocol of the British Hypertension Society (O'Brien et al. 1993) are widely accepted, and most commonly used in publications (see Figure 4), although similar protocols exist in Germany and USA SP10 (AAMI 2007). In addition, CEN standards including clinical validation protocols are available for the manufacturers to use (EN 1060-1, 2 and 3, CEN 1995a, 1995b, 1997). In the recent years there has been a steady increase in the clinical validation of blood pressure measurement devices (see Figure 4). All clinical validation protocols require the use of Hg sphygmomanometers as reference but the CEN standards also allow the use of alternative measurement devices.
Figure 4 Cumulative graph of validation studies performed according to the European Society of Hypertension International Protocol (ESH-IP) compared to the British Society of Hypertension (BHS) and the US Association for the Advancement of Medical Instrumentation (AAMI) protocols from 2002 until June 2009 (Modified from Stergiou et al. 2009b).
The clinical validation protocols presented on Figure 4 require a series of consecutive blood pressure measurements taken over a wide range of blood pressures using the test device in comparison to the mercury sphygmomanometer as a reference. The accuracy of the test device is graded (A-D – where A or B is a pass) or given a pass/fail for systolic and diastolic pressure accuracy according to each protocol. This is usually based on the number/percentage of differences between observer and device in three categories: differences ≤5mmHg, ≤10mmHg and ≤15mmHg. In addition the mean difference and standard deviation (SD) of the difference is calculated and measured against the ANSI/AAMI SP10-1992 standard (AAMI 2007), which requires a mean difference (SD) ≤5 (8) mmHg for clinical recommendation. Devices that have been assessed according to these standards are subsequently listed on the British Hypertension Society and other websites after independent review by the respective committee members of these organisations who give a final verdict as to whether the device should be recommended for clinical use or not, based on whether the protocol guidelines were adequately followed.
Despite the concern that the majority of devices have not yet been validated, it is encouraging to note that the number of validation studies has steadily risen from only 10 in 1990 to 104 studies in 2009 [Stergiou et al 2009b). The British Hypertension Society and other websites are valuable resources for both clinicians and patients.
Mercury is toxic, and there exists the Community Strategy Concerning Mercury with the aim of restricting the use of mercury. Mercury sphygmomanometers have been instrumental in developing the present knowledge on hypertension as a risk factor for cardiovascular diseases and its control by treatment. Therefore, they are considered the gold standard for blood pressure measurement. The need for accurate clinical measurement will always be present, and the fact that important clinical decisions will continue to be made on very small numbers of readings (often one, and rarely more than three) emphasizes the need for maximum accuracy.
Several aneroid and automated alternative blood pressure devices have been validated against the mercury sphygmomanometer. Currently there are no reports published on any electronic device that has been validated using aneroid machines. It can be envisioned that in the future one of the alternative blood measurement devices might also be suitable as a reference for clinical validation of newly developed devices. Until a suitable mercury-free device is developed and recognised as a reference for blood pressure measurement, mercury sphygmomanometers will be needed for clinical validation studies of aneroid and automated blood pressure measurement devices.
It is recommended that for clinical validation studies mercury sphygmomanometers should remain available as reference for alternative mercury-free blood pressure measurement devices.