Accurate Blood Pressure Measurement
For Medical Instrumentation


Design of a Blood Pressure Monitor

There are many factors that go into the design of a blood pressure monitor depending on market needs. There are usually tradeoffs involving cost, measurement accuracy, measurement speed, ease of use, subject age range, device size, and artifact rejection. While all devices require consideration of most if not all of these factors, different markets require more attention to some of these variables.

A SpaceLabs Ambulatory BP MonitorHome blood pressure monitors require accuracy, ease of use, and low cost. Ambulatory monitors require accuracy, comfort, small size, and resistance to artifacts. Clinical monitors require accuracy and low cost. Critical care monitors require reasonably accurate measurements quickly obtained even if the subject is shivering. They also frequently require that they be suitable for use on children as well as adults. An ISO safety standard requires any automatic cycling device to limit maximum pressure and maximum inflation time to certain limits depending on the age of the patient. In addition these limits must be met even in the presence of a single fault. 

It is relatively easy to design a monitor that will work well on a subject that is sitting still, has a regular heartbeat, and is motion free, as is usually the case in clinical or home measurements. It is more difficult to design a monitor that will quickly capture a subject's blood pressure when the subject is moving or is being transported. Many of the steps that can be taken to improve accuracy also extend the measurement time.

Low cost digital circuitry has made the hardware design relatively simple. Even most adjustments (pots) can be eliminated, greatly reducing the cost of calibration. The challenge comes in selecting the measurement method, and designing the firmware to provide the performance required.

Sources of Artifact and Arrhythmia
in Automatic Measurement

1. Artifact

a. Muscle activity
b. Tendon movement
c. Respiration
d. Vibration such as that caused by transportation
e. Shivering
f.  Elevation changes

2. Arrhythmia

a. Missed beats (Algorithm must accommodate)
b. PVC (Algorithm must accommodate)
c. PAC
d. Atrial Fibrillation
e. Ventricular fibrillation (ignore. BP not useful)
f.  Erratic heart beats (Non-sinus rhythm)

A good strategy is to attack artifact and arrhythmia at five levels.

1. Pre-processing to eliminate artifact before collection
2. Pause at a pressure level until enough information is collected if artifact is present
3. Pulse qualification scheme
4. Methods used to evaluate blood pressure "table"
5. Tests of sensibility after the best attempt to obtain BP

The basic measurement scheme to a large extent determines how well the device will function. There are three different schemes. Each has its own advantages and disadvantages.

1. Measure during inflation
2. Measure during step-wise deflation
3. Measure during linear deflation

The design is part art and part science. It usually takes some experimentation before acceptable performance is achieved, especially if the equipment is to be artifact tolerant.

Measurement Speed

In general, the faster the measurement requirement, the more difficult it is to obtain an accurate reading. Fortunately applications that require a fast measurement usually do not have to be as precise as applications that do not require a fast measurement. Perhaps the most demanding for accuracy while being subjected to considerable artifact is an ambulatory blood pressure device. Such a device is normally worn by a subject for 24 hours, and measures blood pressure several times an hour while the subject engages in their normal activities. The author has extensive experience in the design of the world's most popular ambulatory blood pressure monitors (90202, 90207, 90217).

Safety Considerations

A blood pressure monitor normally obtains the blood pressure by means of an occluding cuff. International regulations require that an automatic cycling device be single-fault-tolerant to prevent too long of inflation, or inflation to a pressure that is too high. These time and pressure limits differ between adults and children. While the author has designed systems that use a single processor and meet these requirements, the easiest means is to use two processors in the control of the device.

Clinical Validation

One of the surprises we encountered early in our blood pressure career was the need to teach health care professionals how to correctly measure blood pressure. The company I was working for was designing products in Oregon and testing them at a well-known Eastern university. Our initial results were poor, and changes in the algorithm were not successful at resolving differences. It was only when we started using two "experts" measuring blood pressure at the same time (using a teaching stethoscope) that we located a major source of our problem. The experts could not agree on the subject's blood pressure. So we had to get really good at teaching the experts how to measure pressure. The point of this discussion is that the design of the clinical testing is every bit as important as the design of the device.  Also, the monitoring of the clinical validation by an expert will usually catch lapses in following the protocol.

Blood Pressure Cuffs

Blood pressure cuffs play a very important role in the indirect determination of blood pressure, yet they are usually not well understood or properly used. The size of the cuff (at least the inflatable portion, or bladder) and the application of the cuff are of great importance in obtaining an accurate measurement, yet this size is frequently ignored. I have visited the clinics of luminaries in the blood pressure field, and have asked them what size bladder they use. Frequently they do not know. It is commonplace to find that the size depends on the examination room, as different brands of cuffs are employed, each being differently sized. Yet the size of the cuff will greatly influence the measurement results. Also, the way the cuff is applied to the limb greatly influences the results. It is estimated that at least 30% of blood pressure measurements are incorrect because of cuff selection or application (Review: A century of confusion; which bladder for accurate blood pressure measurement? O'Brien J. Hum Hypertens 1996; 10:565-572). The most frequent error is in using a cuff that is too small, which results in blood pressure measurement that is too high, resulting in millions of people in the US alone being placed on medication that they do not need.

Even when the "correct" cuff is selected there is still a significant difference in blood pressure measurement if the limb size is in the range of two cuffs. Using the larger size cuff will produce a reading significantly lower than using the smaller size.

There are two schools of thought regarding the cuff. One believes the larger the cuff (inflatable portion) the better, while the other school believes too large of a cuff produces errors similar to a cuff that is too small.

The American Heart Association (AHA makes recommendations about the cuff size, although it is not clear. It appears that the battle between the two schools continues to spill over into their recommendations.

Cuff size has been the topic of discussion for about 100 years. The first cuff was a bicycle innertube width affair suitable for use only on the smallest of arms. The next commonly-used size was about 12 cm in width, similar to the most commonly-used cuff size today. Length has varied from 50% to 100% of arm circumference. If the function of the cuff is to apply the same pressure to the artery as is in the cuff, then as postulated by Mike Cohen of CAS Medical (now deceased), the bigger the better. The limb presents a back resistance to the cuff trying to compress it. The maximum pressure point is in the center of the cuff, and decreases towards the edges. The wider the cuff, the closer the pressure exerted on the artery approaches the pressure in the cuff. 

Studies show that varying the width from the nominal changes the blood pressure reading. Too wide of a cuff results in too low of a reading, and too narrow results in an artificially high reading. The reason a wide cuff produces too low of a reading is not obvious. Various researchers first did not believe it, because it does not seem to make sense, and then begrudgingly accepted it as true. The author believes an occluding cuff applied to the limb prevents the reflection that adds to the incident wave. This is referred to as "systolic heightening." This would result in too low of a reading, unless compensation was provided. The narrower cuff provides a first-order compensation.

For diastole, a similar compensation is occurring, but this time to compensate for using Korotkoff phase V instead of phase IV. This explanation is unknown in the medical field. You will only learn of it here.

Types of Cuffs

There are two different types of cuffs in common use. One type has a two-piece construction consisting of a cloth wrap that contains a rubber bladder. The bladder will blow up much the same as a balloon if not contained by the cloth wrap. This will cause inaccuracies if the pocket holding the bladder is not sized to closely limit any bladder expansion. Similar to a balloon, the bladder will retain its shape and will be resistant to expanding its volume until the first time it has been over- inflated. This can occur if the cuff is attached to an automatic device and the cuff is inflated without it being wrapped around an arm. Once this has occurred, the bladder may not be useable, unless the cuff pocket prevents any further expansion when the cuff is used. This source of inaccuracy is exacerbated if the cuff is loosely applied to the arm.

The other type of cuff is a one-piece design where the bladder is formed as part of the wrap. This bladder is non-distensible, and the author has found that it will result in a lower reading even if it is the same size as a cuff with a rubber bladder. Because the bladder is non-distensible, the application technique is less important for the one-piece style of cuff.

Cuff Tightness

It has long been known that it is important to apply the cuff snuggly. This is because the space between the limb and the cuff must be filled by the bladder before pressure is applied to the limb. If the space is large, then the bladder inflation causes the bladder to narrow where it contacts the arm, acting as a narrower bladder. What is less appreciated is that the length of the bladder influences the degree of narrowing. If the bladder almost fully encircles the arm, then it does not narrow nearly as much as a bladder does when it is only partially encircling the arm. Thus the bladder length is not really important if the cuff is applied correctly, and becomes more important as the cuff is applied loosely. Most cuffs are designed such that the bladder fully encircles the smallest arms and encircles less than 70% of the largest arms within the specified arm range. In general, the longer the bladder the better.

Cuff Centering

Cuffs are marked such that the center of the bladder is positioned over the artery if the cuff is correctly applied. A little analysis will show the pressure applied by the cuff to the limb is the same all of the way around, except where the transition occurs at the ends of the bladder. At this point, there is an area where the inflated bladder prevents the adjacent cuff material from contacting the arm. If this area is over the artery, considerably less pressure is applied resulting in a falsely high reading.

The Dilemma of the Tapered Arm

Most arms exhibit a taper. That is, the arm near the shoulder is larger than the arm near the elbow. Most practitioners apply the cuff in a cylindrical manner such that the cuff is snug at the upper part of the arm, and loose at the bottom edge. This will result in a narrowing of the bladder as the bladder is inflated. The amount of narrowing depends on the degree of taper, and the length of the bladder. Usually it is better to wrap the cuff such that it is of equal tightness at both the top and the bottom. This results in less inflation needed. However, it also effectively results in a narrower bladder, because the cuff is no longer symmetrical around the arm. This narrowing is not as severe as is produced in the former case of cylindrically wrapping the cuff.

New Development in Cuffs

The author has invented a "one size fits all" cuff.  This cuff was used on the Pharma-Smart blood pressure device, and the clinical results are detailed in Blood Pressure Monitoring 2004, 9:19-23. The cuff has the added advantage of being quite difficult to misapply, and should eliminate most of the errors produced by cuff selection and application at a low cost. A patent is pending. More information will be provided to qualified interested parties. After the patent is issued, we will disclose all of the advantages of this new approach