[Index FAC] [FVCC Index]
Hipertensión Arterial/Hypertension
Performance of the COLSON MAM BP 3AA1-2
Automatic Blood Pressure Monitor according to
the European Society of Hypertension
Validation Protocol

Pereira T., Maldonado J.

Instituto de Investigação e Formação Cardiovascular,
Escola Superior de Tecnologias da Saúde de Coimbra, Portugal.


Objective: Evaluate the performance of the oscillometric automated blood pressure monitor Colson MAM BP 3AA1-2 according to the validation protocol of the European Society of Hypertension, testing its adequability to self-measurement of blood pressure. The behavior of the device was assessed in relation to several clinical variables, like age, gender, body mass index, arm circumference and arterial distensibility.
Method: 33 subjects (15 men and 18 women) with a mean age of 47±10 years were included and analyzed according to the procedures included in the European Society of Hypertension validation protocol. Same arm sequential blood pressure measurements were made, alternating between a mercury standard and the automatic device. The differences among the test-control measurements were assessed and distributed by categorization zones of 5, 10 and 15 mmHg of discrepancy. Aortic pulse wave velocity was assessed to all subjects with a Complior device (Colson, Paris).
Results: The Colson MAM BP 3AA1-2 passed all three phases of the protocol for both systolic and diastolic blood pressure. The mean difference between the test and control measurements were -1,0±5,0 mmHg for systolic blood pressure and -1,1±4,1 mmHg for diastolic blood pressure. Both standard deviations are well below the 8 mmHg requisite proposed by the Association for the Advancement of Medical Instrumentation. The predictive value of several clinical variables for the discrepancies was assessed by a regression model analysis, having found none variable that independently undermined the performance of the monitor. In another regression analysis we found a similar relation between test and control blood pressures and a widely recognized and validated index of target-organ damage, the aortic pulse wave velocity.
Conclusions: These data show that the Colson MAM BP 3AA1-2 satisfies the quality requisites proposed by the European Society of Hypertension, demonstrating its adequacy to the inclusion in integrated programs of clinical surveillance based on the self-measurement of blood pressure. The uniformity of its performance among a wide spectrum of clinical characteristics and the relation found with pulse wave velocity further reinforce its clinical validity.



Self-measurement of blood pressure is an important and actual scientific topic, which, although not yet largely supported by experimental evidences, is conceptually embedded with an enormous potential, being situated presumably in a middle position between office blood pressure (OBP) and ambulatory blood pressure monitoring (ABPM) [1-4], regarding the fact that it provides a considerable amount of blood pressure measurements in ideal conditions with a financial impact close to the OBP one [5]. In addition, this method generates a greater involvement of the patient with his disease which may reinforce his individual conscience motivating an adequate compliance to the therapeutic prescriptions.

In self-measurement method, the use of automatic devices makes all sense, given the fact that these simplify the patient's task and limit the potential errors. Nonetheless, given the large number of monitors available, it's indispensable to control the quality of their performance as well as their adaptation to the consumer [6].

Recently the European Society of Hypertension (ESH) developed a new protocol to the validation of automatic blood pressure devices, conjugating the fundamental principles of preexisting protocols [7,8] but simplifying the procedures, making the validation process methodologically simpler and equally trustworthy [9,10].

This paper reports on the accuracy of the automatic blood pressure monitor Colson MAM BP 3AA1-2, evaluated according to the ESH protocol, regarding its validation to the self-measurement of blood pressure.

Thirty-three subjects (15 men e 18 women) with the range of blood pressure required by the ESH [9] were included (Table 1). Mean age ± SD was 47 ± 10 years (range 32 71 years), systolic blood pressure (SBP) 142 ± 27 mmHg (range 100 180 mmHg) and diastolic blood pressure (DBP) 88 ± 16 mmHg (range 61 120 mmHg). Body mass index and arm circumference were respectively 26 ± 5 Kg/m2 (range 18 33 Kg/m2 ) and 28 ± 3 cm (range 24 32 cm). Other clinically relevant features evaluated were antihypertensive treatment, physical activity, smoking status and personal history.

All subjects agreed to participate in the protocol and gave their informed consent.

Table 1: Subjects distribution by blood pressure class

The validation team was composed by two observers and one supervisor. All the elements of the team had wide experience in blood pressure measurement, and the supervisor was an expert on hypertension. Observer training was undertaken prior to the validation process.

The protocol was initiated after a 10 min resting period by each subject. Sequential same-arm measurements were performed as recommended by the ESH [9] (table 2).

Table 2: Sequence of blood pressure measurements during the validation test

The observers took the blood pressure measurements on the right arm with a mercury sphygmomanometer and adult medium cuff, whose bladder had to cover 80% of the arm circumference. The measures obtained were compared and discrepancies of less then 4 mmHg were accepted, otherwise the measurement was repeated.

The mean value of the first measurement performed by the observers was used to allocate the subject to the respective blood pressure class (table 1).

The differences between the test device measures and the mean of the observers measures were categorized according to four bands of accuracy, in which band 1 corresponds to differences less than 5mmHg, band 2 differences ranging from 6 to 10 mmHg, band 3 differences ranging from 11 to 15 mmHg and band 4 differences above 15 mmHg. The analysis was based on the evaluation of the cumulative distribution of the band's values on three zones, corresponding zone 1 to the values categorized on band 1, zone 2 to the cumulative values of bands 1 and 2, and zone 3 to the cumulative values of bands 1, 2 and 3.

After the validation protocol, each subject was undertaken a pulse wave velocity (PWV) determination in the aortic territory with the Complior device (Colson, Paris) according to a method perfectly established and internationally validated [11].

The analysis of the discrepancies was performed in two phases, as established by the EHS [9]. The results for the distribution of the differences included in the analysis by the zones of categorization are described in table 3. In the first phase, It was verified that, for SBP, 37, 44 and 45 measures fell into zones 1, 2 and 3 respectively. For DBP, the corresponding figures were 35, 45 and 45, resulting in the approval of phase one in accordance to the requisites of the ESH [9].

As for the second phase, it was found that, for SBP, 76, 93 and 99 measures fell into zones 1, 2 and 3 respectively. For DBP a similar distribution was observed, with 79, 97 and 99 measures distributed by the respective zones of categorization. The second part requires that, in at least twenty two subjects, two measurements of the three comparisons fall into zone 1, and that all measurements fall above zone 1 in no more than three subjects. The corresponding results were of 28 and 1 subjects for SBP, and of 28 and 2 subjects for DBP.

Table 3: Device validation results table

The differences among measurements were expressed graphically in figure 1 as Bland-Altman plots, illustrating that the automatic monitor slightly underestimated the mean of the observers measurements. The mean discrepancy between the automatic device and the mean of the control measures was -1,0 ± 5,0 mmHg for SBP and -1,1 ± 4,1 mmHg for DBP.

Figure 1: Bland-Altman plots of the automatic device-observer differences for systolic (a) and diastolic (b) blood pressures.

In a multiple regression model, we found no association between any of the considered variables (BP, anthropometric data and stiffness) and the absolute or relative differences between the test and the control measurements. Additionally, a regression analysis between the blood pressure parameters and PWV was performed, revealing no significant differences whether the blood pressures were undertaken by the mercury standard or by the automatic device (figure 2).

Figure 2: Relation of PWV to blood pressure estimation by each method


Self-measurement of blood pressure provides multiple readings during a long screening period and in ideal conditions, contributing to an improved individualization of the blood pressure behavior which, given the linear relation between the blood pressure levels and the cardiovascular risk, may contribute to enhance and optimize the clinical approach of the hypertensive patient with a perfectly acceptable financial impact [1-5]. In this sense, the existence of automatic blood pressure devices is vital as they truly allow the clinical profits of the presumable potential of the self-measurement approach. The recourse to this technology results in the facilitation of the operator task in self measuring blood pressure and also in the reduction of potential bias due to measurements memorization, as it as been demonstrated that the credibility of patient's reports of self-measured blood pressure is quite low [12].

Nevertheless, to avoid misjudgments and errors that might themselves constitute sources of bad diagnosis or bad therapeutic decisions, a thorough evaluation of this automatic devices quality its imperative. For the present study we used the validation protocol developed by the ESH [9], to assess the quality of an available automatic blood pressure monitor. The results clearly show that the device provides accurate and reliable blood pressure measurements. The mean difference between the reference measurements and the automatic device was minimal (about 1 mmHg) and the standard deviation of the discrepancies test device-observers (5,0 and 4,1 mmHg, SBP and DBP respectively) accomplished the requisites proposed by the Association for the Advancement of Medical Instrumentation of a standard deviation < 8mmHg [7].

It was also shown that the reliability of the measurements is maintained over a wide spectrum of subjects with different clinical characteristics. We found no relation between any of the several anthropometric indicators, age, gender or blood pressure levels and the precision of the device. Also, the degree of arterial distensibility, expressed by aortic PWV, wasn't also a determinant factor for its performance.

Additionally we seek to find a relation among blood pressure levels, measured by both methods, and aortic PWV, trying to evaluate the concordance in the determination of target-organ damage. In a regression analysis model, having PWV as the dependent variable, we found a strong concordance in the regression coefficients obtained for the control and automatic device blood pressure levels. Hence the blood pressure measured by this monitor strongly relates with an indicator of target-organ damage, and likewise to the relations observed for the reference blood pressure estimates.

In conclusion, the levels of quality achieved associated with the capability of automatically storing measurements give evidence to the adequacy of the Colson MAM BP 3AA1-2 for self-measurement of blood pressure. The blood pressure measurements with this device correlated well with an indicator of target-organ damage and similarly to the control measurement, reinforcing its clinical validity.


  1. Imay Y, Ohkubo T, Tsuji I et al. Prognostic value of ambulatory and home blood pressure measurements in comparision to screening blood pressure measurements: a pilot study in Ohasama. Blood Press Monit 1996; 1(suppl 2):S51-S58
  2. Ohkubo T, Imai Y, Tsuji I et al. Home blood pressure measurement as a stronger predictive power for mortality than does screening blood pressure: a population-based observation in Ohasama. J Hypertens 1998; 16:971-97
  3. Pickering T for an American Society of Hypertension ad hoc panel: Recommendations for the use of home (self) and ambulatory blood pressure monitoring. Am J Hypertens 1995; 9:1-11
  4. Asmar R, Zanchetti A. On behalf of the Organizing Committee and participants. Guidelines for the use of self-blood pressure monitoring: a summary report of the first international consensus conference. J Hypertens 2000; 18:493-508
  5. Soghikian K, Casper SM, Fireman BH, et al. Home blood pressure monitoring. Effect on use of medical services and medical care costs. Med Care 1992; 30:855-65
  6. O'Brien E. Criteria for validation of devices. Blood Press Monit 1999; 4:279-293
  7. Association for the Advancement of Medical Instrumentation. American National Standard: electronic or automated sphygmomanometers. Arlington, VA: AAMI; 1987
  8. O'Brien E, Petrie J, Littler WA, et al. The British Hypertension Society protocol for the evaluation of blood pressure measuring devices. J Hypertens 1993; 11 (suppl 2):S43-S63
  9. O'Brien E, Pickering T, Staessen J, et al. European Society of Hypertension International protocol for validation of blood pressure measuring devices in adults. Blood Press Monit 2002; 7:3-17
  10. O'Brien E, Atkins N. A comparision of the BHS and AAMI protocols for validating blood pressure measuring devices: can the two be reconciled? J Hypertens 1994; 12:1089-1094
  11. Laurent S, Boutouyrie P, Asmar R, et al. Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertens 2001; 37(5):1236-41
  12. Mengden T, Hernandez RM, Beltran B et al. Reliability of reporting self-measured blood pressure values by hypertensive patients. Am J Hypertens 1998; 11:1413-17






December 1st., 2005

Your questions, contributions and commentaries will be
answered by the authors in the Hypertension list.
Please fill in the form and press the "Send".


Dr. Diego Esandi
Scientific Committee

Dra. Silvia Nanfara
Scientific Committee
Prof. Dr. Armando Pacher
Technical-Steering Committee



This company contributed with the Congress:

Laboratorios Roemmers


©1994-2005 CETIFAC - Bioingeniería UNER
Webmaster Updated: 08/17/2005