Uscom Evidence
Uscom maintains a reference of all current evidence supporting Uscom's products.
For a full list of all evidence see All Evidence
Comparison of SphygmoCor and Pulsecor devices for estimation of central blood pressure
Introduction: The clinical performance testing data described in this report was collected as part of the Southall And Brent REvisitied (SABRE) study. SABRE is a follow-up study of the largest tri-ethnic cohort in the UK, the Brent and Southall population-based cohort, consisting of 1,711 South Asians, 801 African Caribbeans and 2,346 White Europeans.
A number of research questions are addressed by the SABRE design. The objective pertinent to this submission is validation of the agreement between the SphygmoCor radial tonometry system and the Pulsecor oscillometric system for pulse wave analysis.
Method: Inclusion criteria in the study were participants who had taken part in the original Southall and Brent study. Participants were invited to take part in the SABRE study, but were excluded on the basis of severe co-morbidity sufficient to preclude a one-day clinic visit, or those unable to provide written informed consent.
SphygmoCor measurements were collected using the SphygmoCor PX system serial number 66-0013-1999 and processed using SphygmoCor 2000 software. Pulsecor measurements were acquired using device serial number R6.5-010.
Two investigators were involved in the collection of data, one being an experienced cardiovascular physiologist, and the other holding PhD in cardiovascular and arterial physiology. Prior to the start of data collection, both were experienced in using SphygmoCor to collect data from the radial and carotid arteries. They were also trained by senior nurses to acquire accurate brachial blood pressure measurements using a cuff and are very familiar with pressure waveforms.
All participants fasted overnight and refrained from taking any medication on the morning of their clinic visit. Participants rested in a sitting position in a quiet, dark temperature-controlled room for at least 10 minutes prior to measurements. All measurements were made on the left brachial and radial arteries. SphygmoCor measurements were acquired immediately after Pulsecor measurements. Cuff size was selected to be appropriate to the participant’s arm size. Available cuff sizes were marked for arm circumferences Large Adult: 33.5 to 46.0 cm, Adult: 27.5 to 36.5 cm, and Small Adult: 20.5 to 28.5 cm.
The difference between SphygmoCor and Pulsecor central systolic, mean, diastolic and augmentation pressure was calculated for each paired measurement. SphygmoCor requires an external source of blood pressure for calibration; brachial cuff systolic and diastolic pressures were used for this purpose. As there is no internationally accepted standard for evaluation of agreement of augmentation index, the SP10 and BHS criteria have been applied to evaluating the augmentation pressure, AP. AP is calculated for both systems using the same formula AP = AIx.PP, where AIx is the augmentation index determined by the system, and PP is central pulse pressure (central systolic – central diastolic pressure). Agreement between SphygmoCor and Pulsecor is assessed on the basis of AAMI SP10 and BHS criteria.
Results: 1438 participants came into the SABRE clinic. 1318 measurements were made with SphygmoCor and 1428 measurements were made with Pulsecor. Paired Pulsecor and SphygmoCor measurements were available in a total of 1161 individuals. Paired measurements were excluded if the signal to noise ratio reported by Pulsecor was < 3 dB (n=7), peripheral augmentation index was reported as <0 (n=7), or results were unable to be calculated by the SphygmoCor software (n=6) or Pulsecor algorithm (n=4). 1139 paired measurements were available after these exclusions.
Demographic information about the study population who had paired measurements is shown in Table 1.
Discussion: ANSI/AAMI SP10:2002/A2:2006/(R)2008 protocol Method 1 requires no fewer than 85 subjects and a minimum of 255 observations. The mean difference between test and comparison system should be ±5 mmHg or less, and the standard deviation must be ±8 mmHg or less. Based on the results of this investigation, the Pulsecor system comfortably meets these criteria for central systolic, diastolic, mean and augmentation pressures.
According to the British Hypertension Society grading criteria (O’Brien, The British Hypertension Society protocol for the evaluation of blood pressure measuring devices, Journal of Hypertension 1993, 11 suppl 2:S43-S62), the percentage of differences no more than 5, 10 and 15 mmHg should be greater than 60%, 85% and 95% respectively to achieve a Grade A. Based on the results of this investigation, the Pulsecor system is rated Grade A for systolic, diastolic, mean and augmentation pressures.
There were limited numbers of subjects presenting with low blood pressures (i.e. systolic < 100 mmHg, diastolic < 60 mmHg) in this cohort. A cohort including participants with lower blood pressures has been studied separately with very equivalent results.
The Pulsecor and SphygmoCor systems for non-invasive estimation of central systolic, diastolic and mean blood pressures have substantially equivalent overall system efficacy when used on adult patients. The systems also provide substantially equivalent augmentation index and augmentation pressure.