nursing research july/august 2006 vol 55, no 4, 283–291 · hemodynamics relatively unobtrusively...

Ambulatory Impedance CardiographyA Systematic Review

Monica J. E. Parry 4 Judith McFetridge-Durdle

b Background: Standard noninvasive impedance cardiography

has been used to examine the cardiovascular responses

of individuals to a wide range of stimuli in critical care and

laboratory settings. It has been shown to be a reliable

alternative to invasive thermodilution techniques and an

acceptable alternative to the use of a pulmonary artery

catheter. Ambulatory impedance cardiography provides a

similar assessment of cardiac function to standard non-

invasive impedance cardiography, but it does so while

individuals engage in activities of daily living. It offers

portability and the option of managing complex patients in

outpatient settings.

b Objective: To critically examine through a literature analysis thevalidity, reliability, and sensitivity of ambulatory impedance

cardiography for the assessment of cardiac performance

during activities of daily living.

b Methods: The Cochrane Database of Systematic Reviews

(CDSR), The Cochrane Database of Methodology Reviews

(CDMR), The Cochrane Central Register of Controlled

Trials (CENTRAL), Database of Abstracts of Reviews of

Effects (DARE), National Health Service Economic Evalu-

ation Database (NHS EED), Health Technology Assess-

ment (HTA), and The Cochrane Methodology Register

(CMR; 1966Y2005); MEDLINE (1950Y2005); and CINAHL

(1982Y2005) were searched using the following terms:

ambulatory cardiac performance, impedance cardiac per-

formance, AIM cardiac performance monitor, thoracic

electrical bio-impedance, impedance cardiography, ambu-

latory impedance monitor, bio-impedance technology,

ambulatory impedance cardiography, bio-electric imped-

ance; also included were reference lists of retrieved

articles. Studies were selected if they used an ambulatory

impedance monitor to examine one or more of the

following cardiovascular responses: pre-ejection period

(PEP), left ventricular ejection time (LVET), stroke volume

(SV), or a combination of these.

b Results: Studies have been predominantly descriptive and

have been focused on a young, male population with a

normal body mass index (BMI; 25Y29 kg/m2). Inconsisten-

cies in determining specific markers of cardiac function

(e.g., PEP and SV) across studies necessitated that results

be reported by outcome for each study separately.

b Discussion: Ambulatory impedance monitors are valid and

reliable instruments used for the physiologic measurement

of cardiac performance. Sensitivity is established utilizing

within-individual measurements of relative change. This is

especially important in light of an aging population and

technical advances in healthcare. Further research is

warranted using nursing interventions that focus on an

older, female population who have a BMI greater than

30 kg/m2. Availability of noninvasive ambulatory measures

of cardiac function has the potential to improve care for

a variety of patient populations, including those with

hypertension, heart failure, pain, anxiety, and depressive

symptoms.

b Key Words: ambulatory monitoring & impedance cardiography

Impedance cardiography (ICG) is a noninvasive tech-nology that provides information regarding hemody-

namic and fluid status (Lasater & Von Rueden, 2003).Noninvasive assessments of cardiac performance usingbioelectrical impedance were first used in 1966 (Kubicek,Karnegis, Patterson, Witsoe, & Mattson, 1966) when im-pedance measurements within the thorax were used toestimate cardiac output (CO). Impedance changes aregenerated by fluctuations in blood volume and flow

Nursing Research.July/August 2006.Vol 55, No 4, 283–291

Nursing Research July/August 2006 Vol 55, No 4 283

Monica J. E. Parry, RN, PhD(C), ACNP, CCN(C), is AdvancedPractice Nurse/Acute Care Nurse Practitioner, Cardiac Surgery,Kingston General Hospital, Kingston, Ontario, Canada; DoctoralCandidate, Faculty of Nursing, University of Toronto; andStrategic Training Fellow in the FUTURE Program for Cardio-vascular Nurse Scientists.Judith McFetridge-Durdle, PhD, RN, is Associate Professor,School of Nursing, Dalhousie University, Halifax, Nova Scotia,Canada and Key Mentor in the FUTURE Program for Cardio-vascular Nurse Scientists.

Copyr ight © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

velocity in the ascending aorta during systole and diastole.Impedance to electrical current decreases during systoledue to increased blood volume and flow velocity, andincreases during diastole as flow is reduced. Pulsatile im-pedance changes reflect ascending aortic flow and leftventricular function. Baseline thoracic impedance (Z0),pulsatile impedance/time changes (dZ/dt), and electrocar-diography (ECG) are used to calculate various measuresof cardiac function. Electrocardiographic and impedance-generated hemodynamic waveforms are depicted in Figure1, and impedance-generated hemodynamic parameters anddefinitions are listed in Table 1.

Impedance cardiography is a reliable and valid noninva-sive technique for measuring various indices of cardiovas-cular function in critical care environments and laboratorysettings (McFetridge & Sherwood, 1999; Shoemaker et al.,1996, 1998, 2001). It has been shown to be a reliablealternative to invasive thermodilution techniques and anacceptable alternative to the standard use of a pulmonaryartery catheter in a variety of populations (Shoemakeret al., 1996, 1998, 2001; Van De Water, Miller, Vogel,Mount, & Dalton, 2003), including critically injured obese(r = .85, p G .0001) and nonobese patients (r = .82, p G.0001; Brown, Martin et al., 2005), patients with athero-sclerotic and rigid thoracic aortas ages 55Y70 years (r = .87)and over 70 years (r = .80; Brown, Shoemaker, Wo, Chan,& Demetriades, 2005), patients admitted to the emergencydepartment with cerebrovascular accidents (Velmahos, Wo,Demetriades, Bishop, & Shoemaker, 1998), and hospitalizedpatients with advanced, decompensated chronic heart fail-ure (Albert, Hail, Li, & Young, 2004).

Impedance cardiography has been used also to evaluatethe hemodynamic adjustments underlying blood pressure(BP) responses during mental stress. Overall, evidence hasshown that systemic vascular resistance (SVR) control ofBP during mental stress is a marker of cardiovascular dis-ease risk (Light & Sherwood, 1989; Sherwood & Turner,1995). Exaggerated SVR response during mental stresshas been associated with myocardial ischemia (Blumenthalet al., 1995). Moreover, SVR responses to stress are bluntedduring the high estrogen phase of the menstrual cycle(McFetridge & Sherwood, 2000). These laboratory find-ings suggest that stress alters hemodynamics and influ-

ences the balance between myocardial oxygen demandand supply.

There are a number of commercially available impedancecardiographs but the Minnesota Impedance CardiographModel 304B (Instrumentation for Medicine, Greenwich,CT) is the most widely validated standard and has beenextensively used in research applications (McFetridge &Sherwood, 1999). It has been used in populations includ-ing (a) normotensive males with and without a family his-tory of hypertension (Hamer, Jones, & Boutcher, 2006);(b) female alcoholics with transitory hypertension afterearly abstinence (Bernardy, King, & Lovallo, 2003); (c) lonelyand nonlonely undergraduate students (Cacioppo et al.,2002); (d) individuals with paraplegia (Raymond, Davis, &van der Plas, 2002); and (e) men with mild, stable heartfailure (Davis et al., 2006). The Minnesota ImpedanceCardiograph Model 304B has also been used to (a) inves-tigate the effect of sleep on cardiac activity (Carrington,Walsh, Stambas, Kleiman, & Trinder, 2003), (b) determineracial differences in vascular reactivity (Kelsey, Alpert,Patterson, & Barnard, 2000), and (c) examine the effects ofsmoking and oral contraceptive use on the hemodynamicresponses to stress in women (Straneva, Hinderliter, Wells,Lenahan, & Girdler, 2000). This model uses a 4-mA con-stant current source with a 100-kHz oscillator frequencyand includes a display of Z0. A tetrapolar band electrodeconfiguration has been adopted in most studies using the

FIGURE 1. Electrocardiographic and impedance-generatedhemodynamic waveforms.

qTABLE 1. Impedance-GeneratedHemodynamic Parameters and Definitions

HemodynamicVariable Parameter Definition

Thoracicfluid status

Z0 = thoracicimpedance

Baseline fluid statusin chest

Left ventricularfunction

CO = cardiacoutput

Amount of blood ejectedfrom the left ventriclein 1 minute

CI = cardiacindex

Cardiac output dividedby body surface area

Preload SV = strokevolume

Amount of blood ejectedwith each beat

Afterload SVR =systemicvascularresistance

Amount of resistancethat the heart mustpump against

Contractility dZ/dt =impedancechangesover time

Reflects the force ofventricular contraction

PEP =pre-ejectionperiod

Time from ventriculardepolarization toventricular ejection

LVET = leftventricularejection time

Period of time over whichblood is ejected fromthe left ventricle

284 Ambulatory Impedance Cardiography Nursing Research July/August 2006 Vol 55, No 4


impedance technique for monitoring cardiovascular hemo-dynamics. Two electrodes initiate a high-frequency excita-tion current while the impedance is measured across aninner two-voltage electrode. The upper and lower voltageelectrodes are placed around the base of the neck andaround the thorax at the level of the xiphisternal junction(McFetridge & Sherwood, 1999). The outer currentelectrodes are placed at a minimum distance of 3 cm fromthe voltage electrodes (Figure 2).

Impedance cardiography is ideally suited to nursingstudies because it poses minimal risk and is cost-effective(McFetridge & Sherwood, 1999). There is a small risk oflocal skin irritation associated with electrode application.Applying an over-the-counter nonallergenic protectivesubstance to the skin before applying the electrodes reducesthis risk. With the introduction of the ambulatory imped-ance monitor, it should be possible to assess cardiovascularhemodynamics relatively unobtrusively during a widerange of procedures, in response to a variety of interven-tions, and in both static and active environments. Ambu-latory impedance monitors offer portability, so thatassessments of cardiovascular hemodynamics are not re-quired to be in a laboratory or critical care setting. Thistechnology is particularly exciting when the possibilities ofassessing cardiac function, including CO, stroke volume(SV), pre-ejection periods (PEP), and left ventricularejection times (LVET) during activities of daily living(ADLs) are considered. It may be possible to evaluatecardiac performance and, subsequently, manage patients inoutpatient settings rather than admitting patients to thehospital for assessment of their hemodynamic parameters.

ObjectiveThe purpose of this review is to determine whether ambula-tory impedance monitors are valid and reliable instruments,with adequate sensitivity to detect change in cardiac perfor-mance, for use in nursing research and practice.

Methods

Criteria for Considering StudiesThere was no restriction to the type of study or description ofparticipant considered for this review. However, to establishthe power of an ambulatory impedance monitor to assessvarious cardiovascular responses, it was necessary to includein the review studies that compared an ambulatory imped-ance monitor to a reference standard. It was also necessary torestrict the instrumentation, or the type of an ambulatory im-pedance monitor, to one that examined one or more of thefollowing cardiovascular outcomes: PEP, LVET, and SV.Studies were not included if they did not use an ambulatoryimpedance monitor for the assessment of cardiac performance.

Search StrategyThe Cochrane Library (1966Y2005), including the CochraneDatabase of Systematic Reviews (CDSR), The Cochrane Data-base of Methodology Reviews (CDMR), The CochraneCentral Register of Controlled Trials (CENTRAL), Data-base of Abstracts of Reviews of Effects (DARE), NationalHealth Service Economic Evaluation Database (NHS EED),Health Technology Assessment (HTA), and The CochraneMethodology Register (CMR); MEDLINE (1950Y2005);and CINAHL (1982Y2005) were searched using the fol-lowing terms: ambulatory cardiac performance, impedancecardiac performance, AIM cardiac performance monitor,thoracic electrical bio-impedance, impedance cardiography,ambulatory impedance monitor, bio-impedance technol-ogy, ambulatory impedance cardiography, and bio-electricimpedance (Deville, Bezemer, & Bouter, 2000; Farbey,1993; Greenhalgh, 1997; Haynes, Wilczynski, McKibbon,Walker, & Sinclair, 1994; Higgins & Green, 2005).

MEDLINE was searched using a combination of textwords and Medical Subject Headings (MeSH; 1950Y2005;Lowe & Barnett, 1994). Using the MeSH database, thefollowing MeSH headings were established: cardiography,

FIGURE 2. Typical band electrode placement for the Minnesota Impedance Cardiograph Model 304B (Instrumentation for Medicine, Greenwich, CT)[reference standard].

Nursing Research July/August 2006 Vol 55, No 4 Ambulatory Impedance Cardiography 285


impedance and monitoring, ambulatory. The MeSHheadings were combined and the scope of the search wasrestricted to the major topic headings, including impedanceplethysmography and transthoracic impedance.

Lastly, the reference lists of all retrieved papers wereexamined. Included studies were published in English.Dissertations and conference proceedings were excludedfrom the review.

Methods of the ReviewAll publications identified were evaluated to ensure theymet the inclusion criteria. Standard methods were used tocollect data and assess the methodological quality of thestudies (Higgins & Green, 2005; The Evidence-BasedMedicine Working Group, 2002). Due to the clinicaldiversity of the studies included in this review, a meta-analysis was not done. Inferences about the results of thestudies are based on a critical appraisal, conducted by theauthors, and not on meta-analytic techniques.

Description of StudiesSeven studies were identified in the search (Barnes,Johnson, & Treiber, 2004; Bishop, Pek, & Ngau, 2005;Hawkley, Burleson, Bernston, & Cacioppo, 2003; Rieseet al., 2003; Sherwood, Hughes, & McFetridge, 2003;Sherwood, McFetridge, & Hutcheson, 1998; Vrijkotte, vanDoornen, & de Geus, 2004). Reference lists of these studiesrevealed another four studies for examination (Boomsmaet al., 2000; Kizakevich et al., 2000; Nakonezny et al.,2001; Vrijkotte, van Doornen, & de Geus, 2000). Whenexamined, two studies were excluded: the results were notyet available in one study (Boomsma et al., 2000), and itwas difficult to determine the type of impedance monitorused in the second study (Kizakevich et al., 2000).

Of the nine studies meeting the inclusion criteria, in twothe validity, reliability, and sensitivity of an ambulatoryimpedance cardiograph had been evaluated (Sherwoodet al., 1998, Nakonezny et al., 2001) with the most widelyvalidated standard, the Minnesota 304B. One study was arandomized controlled trial comparing the impact of dailyhealth education on lowering BP in African-Americanadolescents (Barnes et al., 2004). In one study, the fea-sibility of large-scale ensemble averaging of ambulatoryimpedance cardiograms was examined using the VrijeUniversiteit-Ambulatory Monitoring System (VU-AMS;Riese et al., 2003). The remaining five studies examinedwere the relationship of ethnicity (Bishop et al., 2005;Sherwood et al., 2003), sex and trait anger (Bishop et al.,2005), work stress (Vrijkotte et al., 2000, 2004), andloneliness (Hawkley et al., 2003) on a number of indices ofcardiac function.

Results

InstrumentsThree ambulatory impedance monitors were used toexamine cardiovascular indices: the AIM-8 (Bio-impedanceTechnology, Chapel Hill, NC), the VU-AMS (Version 4.6,VU-FPP, Amsterdam), and the AZCG (World WideMedical Instruments, Dallas, TX).

Ambulatory ImpedanceMonitor TheAIM-8 is amicrocomputer-based bioelectric impedance monitor and signal processing sys-tem designed to assess a number of indices of cardiac functionin a 24-hour ambulatory environment. It consists of a 3 � 4 �1.5 in. plastic enclosure that contains a credit-card-sized bio-electric impedance cardiograph, a credit-card-sized internalcomputer, and a 9-V battery power source. The AIM generatesan 80-kHz, 2-mA constant sine wave alternating current. TheAIM computer section ensemble averages, analyzes, and storesECG, dZ/dt, and Z0 waveforms and the computed cardiacfunction indices during each measurement sequence. The AIMcan function in a continuous or continuous-manual mode andis capable of being activated by a cuff-pressure sensor initiationof each ambulatory BP measurement.

A tetrapolar combination of spot and band electrodeswas developed for use with the AIM system. The recordingelectrodes are band electrodes placed around the base ofthe neck and around the thorax at the tip of the xiphoidprocess, identical to that for the Minnesota referenceconfiguration (Sherwood et al., 1998). Three spot elec-trodes are used as current electrodes, one applied behindthe right ear (over the base of the mastoid process), andthe other two on the lower right and left rib cage 6 cmbelow the lower recording band electrode. The two im-pedance current spot electrodes (right ear and lower ribcage) are sources for the ECG signal, approximating a leadII configuration. The AIM is worn on a belt around thewaist during assessment of normal daily activities. Process-ing of the impedance signals is accomplished using eitherCOPWORKS or COP-WIN (Bio-impedance Technology),which permit ensemble averaging of impedance waveformsto filter noise and respiratory artifacts.

Vrije Universiteit-Ambulatory Monitoring System The VU-AMS is an ambulatory monitoring device designed torecord ECG and ICG recordings from six spot electrodes.Two electrodes on the back send a high-frequency currentand two measuring electrodes on the chest pick up thevoltage drop over the thorax. The lower current electrodeon the back is placed at least 3 cm below the horizontalplane of the lower measuring electrode on the chest, placedat the xiphoid process. The upper current electrode onthe back is placed 3 cm above the horizontal plane of theupper measuring electrode on the chest. The upper chest-measuring electrode is placed at the jugular notch of thesternum between the collar bones. Two additional chestelectrodes are placed, one on each side of the lower chest.Thoracic impedance is assessed against a constant currentof 50 kHz, 0.35mA (Riese et al., 2003).

The VU-AMS does not record the full ECG but insteaduses a continuous time series of R-wave-to-R-wave inter-vals derived from a three-lead ECG (Vrijkotte et al., 2000).The R-wave is used as an approximation of the onset of theelectromechanical systole (EMS) and the PEP scoring ismade relative to the R-wave (Riese et al., 2003). Vagal toneis assessed using a root mean square of successive differ-ences (RMSSD) in these interbeat differences (Vrijkotteet al., 2000). A fixed QYR interval is often added to thisabbreviated PEP to allow easy comparison with the usualQ-wave-based PEP (Riese et al.). Ensemble averaging in-volves digitalizing the ECG and the ICG over all beats



in a preset period of time (e.g., 60 sec) with respect to thepeak of the ECG R-wave, then averaging by summingthe digitalized samples for each signal and dividing by thenumber of synchronized beats (Riese et al.). This procedureis used to reduce the impact of single-beat fluctuations inthe impedance signal related to respiration and thoracicmovements (Riese et al.).

Ambulatory Impedance Cardiograph The AZCG (WorldWide Medical Instruments) is an ambulatory monitor de-signed for noninvasive acquisition of physiological data dur-ing daily activities. It is a 4.5 � 9.5 � 16 cm device weighing400 g with batteries. The analogue subsystem is composedof a three-lead ECG and a four-lead electrical impedancesystem, which produces a constant current source of 50 kHz,2 mA (Nakonezny et al., 2001). The acquired analogue im-pedance signals are filtered, amplified, and differentiated toproduce signals for Z0, $Z, and dZ/dt. The ECG and ICGeach employ a digitally controlled, sampled-signal rebalancemethod for waveform stability. The digital subsystem pro-

vides analogue and digital conversion of signals. Digitalizedsignals are stored on a 20-MB flash card (Nakonezny et al.).Programming during set-up, signal monitoring, and up-loading of data are accomplished using standard communi-cation software through digital input/output connectors anda serial interface to a microcomputer system.

ParticipantsA total of 215 individuals were evaluated with the AIM(Barnes et al., 2004; Bishop et al., 2005; Sherwood et al.,1998, 2003), 197 with the VU-AMS (Riese et al., 2003;Vrijkotte et al., 2000, 2004), and 157 with the AZCG(Hawkley et al., 2003; Nakonezny et al., 2001). The subjects in-cluded in this systematic review were predominantly young,male adults with a normal BMI (25Y29 kg/m2; Table 2).

Validity, Reliability, and SensitivityInconsistency in determining PEP and SV across studiesnecessitated that results be reported by outcome for eachstudy separately.

qTABLE 2. Characteristics of Study Participants by Ambulatory Monitoring System

Study Sample Size (n) Gender Mean Age (years) Ethnicity BMI (kg/m2)

AIM

Sherwood, McFetridge, & Hutcheson (1998) 11 Male (n = 5) 25* White (n = 6) 27

Female (n = 6) African-American (n = 4)

Asian (n = 1)

Sherwood, Hughes, & McFetridge (2003) 20 Male (n = 9) 35* African-American (n = 10) 25

Female (n = 11) White (n = 10)

Barnes, Johnson, & Treiber (2004) 35 Male (n = 21) 16.2 T 1.4 African-American (n = 35) 30

Female (n = 14)

Bishop, Pek, & Ngau (2005) 149 Male (n = 74) 21.5* Chinese (n = 51) N/A

Female (n = 75) Malays (n = 51)

Indian (n = 47)

VU-AMS

Vrijkotte, van Doornen, & de Geus (2000) 109 Male (n = 109) 47.2 T 5.3 Dutch (n = 109) 25

Riese et al. (2003) 21 Male (n = 7) 29 T 5.14 Dutch (n = 21) 22

Female (n = 14)

Vrijkotte, van Doornen, & de Geus (2004) 67 Male (n = 67) 47.1 T 5.2 Dutch (n = 67) 25

AZCG

Nakonezny et al. (2001) 10 Male (n = 10) 24.7 T 2.0 N/A 24

12 Male (n = 6) 20 T 0.52 N/A N/A

Female (n = 6)

Hawkley et al. (2003) 135 Male (n = 68) 19.2 T 1.0 White (n = 112) G27

Female (n = 67) African-American (n = 9)

Asian (n = 10)

Other (n = 4)

Notes. AIM = Ambulatory Impedance Monitor; VU-AMS = Vrije Universiteit-Ambulatory Monitoring System; AZCG = New Ambulatory Impedance Cardiograph;

BMI = body mass index; N/A = data not available.

*SD not recorded.



Validity Validity is the degree to which an instrumentmeasures what it is supposed to measure (Gassert, 1990;Hill, 1988; Portney &Watkins, 1993). Mean values for fourindices of cardiac function measured while subjects weresitting and standing with the AIM compared to theMinnesota 304B (standard) and the AZCG compared tothe ZCG-Minnesota 304B (standard) monitoring systemsare depicted in Figure 3. Cardiovascular responses to sittingand standing are similar across all groups. No data areavailable for posture and monitor effects for the VU-AMS.Pearson’s r values for the correlations between the AIM andthe Minnesota 304B (standard) cardiac function indicesmeasured while the subjects were sitting and standingranged from .87 to .96 (Sherwood et al., 1998). Pearson’sr values for the correlations between the AZCG and ZCG-Minnesota 304B (standard) cardiac function indices againmeasured while the subjects were sitting and standingranged from .68 to .98 (Nakonezny et al., 2001). Twoambulatory impedance systems (AIM, AZCG) correlatewell with the standard Minnesota 304B. No data are avail-able comparing the third ambulatory impedance system(VU-AMS) to a standard.

Reliability Reliability and validity are related intricately.Reliability is the consistency, accuracy, and precision of ameasure (Hill, 1988). With repeated measures of cardiacindices, the less variability the AIM shows, the higher itsreliability. Reliability also reflects accuracy, such that areliable instrument gives the true score and minimizes error.

Sherwood et al. (2003) assessed the hemodynamicvariations in ambulatory BP (ABP) measurements in African-Americans during ADLs. Impedance cardiographic assess-

ments of CO were synchronized to each ABP measurementusing the AIM. They found CO to be a significant predictorof systolic BP in Whites [Z(1, 429) = 3.38, p = .0007] butnot in African-Americans (p = .61), whereas SVR was asignificant predictor of systolic BP in African-Americans[Z(1, 393) = 3.37, p = .0007] but not in Whites (p = .46).Their observations provide preliminary evidence that indi-vidual differences in hemodynamic patterns of BP regula-tion observed in a laboratory environment are reproduciblewith ambulatory impedance measures taken during ADLs.

The AIM was also evaluated in African-Americanadolescents (n = 35) to determine the reproducibility ofdaytime and nighttime ambulatory bioimpedance-derivedmeasures of hemodynamic function (Barnes et al., 2004).Reliability may be defined as the degree of consistency withwhich an instrument measures a physiological variable(Gassert, 1990). It is considered to be synonymous with theaccuracy of the instrument. Across 2 months, Barnes et al.(2004) found heart rate (HR; r = .81) to be highlyrepeatable and SV (r = .54), CO (r = .56), PEP (r = .59),and LVET (r = .74) to be moderately repeatable.

Stroke volume (ml/min) is often determined as per theequation derived by Kubicek et al. (1966): SV = D(L/Z0)

2(LVET)(dZ/dt)max, where D (; cm) is set to a constantvalue of 135; L (cm) is the distance between recordingelectrodes; Z0 (;) is the basal thoracic impedance related toair, blood, and tissue levels; and LVET (msec) is the leftventricular ejection time measured from the B-point to theX-point of the dZ/dt waveform. Using this derivation, andthe AZCG, Hawkley et al. (2003) reported that lonelinesspredicted a lower CO (l per min = HR � SV) during anormal day. They found that differences observed in a

FIGURE 3. Mean values for cardiovascular indices in subjects during sitting and standing postures measured using the AIM monitor compared to theMinnesota 304B (standard) and the AZCG compared to the ZCG-Minnesota 304B (standard). HR = heart rate; SV = stroke volume; PEP = preejectionperiod; LVET = left ventricular ejection time; Si = sitting; St = standing. , AIM; , Minnesota; , AZCG; , ZCG.



laboratory setting generalized to everyday life. Lonelyparticipants exhibited a higher total peripheral resistance(TPR; dyne sec cmj5, a measure compared to SVR) and alower CO than nonlonely participants across the varioussituations and social contexts in which they were measuredduring a normal day.

The reliability of a measurement to detect responsechanges over time is dependent on the reproducibility of themethod (Vinet, Nottin, Lecoq, Guenon, & Obert, 2001).Using the VU-AMS, Riese et al. (2003) found large indi-vidual differences in absolute PEP. They concluded thatmeasuring individual differences in sympathetic activationby using absolute PEP confounds sympathetic drive withindividual differences in adrenoceptor functioning. Forexample, a shorter PEP indicated improved contractility/sympathetic activation and a longer PEP indicated dimin-ished contractility; and increased afterload caused length-ening of PEP and increased preload caused shortening ofPEP (Obrist, Light, James, & Strogatz, 1987). Impedancecardiography usually estimates PEP as the time intervalbetween the ECGYQ wave (onset of EMS) and the B-pointin the ICG, which is the start of the rapid upslope of dZ/dtto its maximum value (Sherwood et al., 1990). The PEP isdefined as the period between the onset of EMS and theonset of left ventricular ejection at the opening of the aorticvalve. However, because the VU-AMS does not record afull ECG, PEP scoring must be made relative to the R-wave, which is used as an approximation of the onset ofthe EMS. The precision and accuracy of an instrument alsoreflects the reliability of the measure. To improve theaccuracy, precision, and reliability, the VU-AMS ambula-tory PEP should be used mainly in a within-subjects design(Riese et al., 2003). Within-subjects changes in PEP acrossthe day can easily be determined via various reactivitymeasuresVlying, sitting, and standing.

Sensitivity For each index of cardiac function, the re-sponse to standing was computed as the difference between

standing and sitting values, expressed as a percent changefrom the sitting value (Sherwood et al., 1998). The meanpostural responses measured using the AIM comparedto the Minnesota 304B and the AZCG compared to theZCG-Minnesota 304B are depicted in Figure 4. TheAIM and AZCG monitoring systems indicate similar pos-tural responses for HR, SV, PEP, and LVET. This wouldsuggest consistency, accuracy, sensitivity, and enhancedreliability.

DiscussionStandard ICG is a reliable and valid noninvasive tech-nique for measuring various indices of cardiovascular func-tion in critical care environments and laboratory settings(McFetridge & Sherwood, 1999; Shoemaker et al., 1996,1998, 2001). It has been shown to be a reliable alternativeto invasive thermodilution techniques and an acceptablealternative to the standard use of a pulmonary arterycatheter in a variety of populations (Shoemaker et al.,1996, 1998, 2001; Van De Water et al., 2003), includingcritically injured patients who are obese (Brown, Martin,et al., 2005), elderly patients with atherosclerotic and rigidthoracic aortas (Brown, Shoemaker, et al., 2005), patientsadmitted to the emergency department with cerebrovascu-lar accidents (Velmahos et al., 1998), and hospitalizedpatients with advanced decompensated chronic heart fail-ure (Albert et al., 2004).

Reliable measurement of cardiac performance andhemodynamic responses during ADLs are extremely im-portant in detecting changes that may be imposed byintervention studies. The validity and reliability of twoambulatory impedance cardiographs (AIM, AZCG) weretested against the reference standard Minnesota 304Bduring sitting and standing. The devices were comparedin healthy subjects. Both the AIM and the AZCG trackedchanges across conditions closely with this reference standardand appeared to provide valid and reliable estimates of

FIGURE 4. Responses to standing, expressedas percent change for standingj sitting values(means) for HR, SV, PEP, and LVET recordedusing the AIM, Minnesota model 304B(standard), AZCG, and ZCG-Minnesota 304B(standard). HR = heart rate; SV = stroke volume;PEP = pre-ejection period; LVET = leftventricular ejection time. , AIM; , Minnesota;, AZCG; , ZCG.



cardiac function (HR, SV, PEP, and LVET). Moreover,both ambulatory impedance systems estimate PEP as thetime interval between the ECG-Q wave (onset of EMS) andthe B-point in the ICG, which is the start of the rapidupslope of dZ/dt to its maximum value. These resultssupport those of Riese et al. (2003) and Sherwood et al.(1990), who recommend that within-individual measure-ments of relative change are more valid than absolutevalues. No data were found describing the validity of theVU-AMS to a reference standard.

In repeated measures of a characteristic, the less vari-ability an instrument shows, the higher its reliability (Hill,1988). Barnes et al. (2004) found HR to be highlyrepeatable and SV, CO, PEP, and LVET to be moderatelyrepeatable across 2 months. However, they did not controlfor postural changes, physical activity levels, or affectivestates in their data analysis, explaining some of thevariability in reproducibility of some of the bioimpedancemeasures. In addition, a number of methodologic factorscan impact reproducibility, such as consistency of electrodeplacement during instrumentation, minimization of elec-trode resistance by thorough preparation of the placementarea (shaving, cleansing, skin prep), and consistency inwaveform editing. Efforts should be made to address theseissues when using ambulatory impedance in outpatientsettings.

Information from this systematic review is importantfor nurses in light of the complexities of the patientpopulations and the technical advances in healthcare. Theclinical utility of ambulatory impedance for the non-invasive monitoring of cardiovascular responses of indi-viduals to various nursing interventions in outpatient settingsis immense. Nurses are ideally positioned to incorporateambulatory ICG into research and practice settings, with avariety of patient populations.

As the population ages, caring for individuals withmultiple comorbid conditions will become a part of every-day practice. Ambulatory ICG provides a similar assess-ment of cardiac function to standard noninvasive ICG, butit does so while individuals engage in ADLs. It offersportability and the option of managing complex patients inoutpatient settings. Further research is warranted compar-ing ambulatory ICG to the reference standard Minnesota304B in older, female populations who have a BMI greaterthan 30 kg/m2.

All studies included in this systematic review weredescriptive. Evaluating the within-individual measurementsof relative change in cardiovascular responses of individu-als to various nursing interventions has the potential toimprove care for a variety of patient populations includingthose with hypertension, heart failure, pain, anxiety, anddepressive symptoms. Because ambulatory ICG permits theexamination of hemodynamic responses to stress duringADLs, it may help identify individuals at risk for futurecardiovascular events. q

Accepted for publication April 6, 2006.Corresponding author: Monica J. E. Parry, RN, PhD(C), ACNP,CCN(C), Cardiac Surgery, Kingston General Hospital, 76 StuartStreet, Kingston, Ontario, Canada (e-mail: [email protected]).

ReferencesAlbert, N. M., Hail, M. D., Li, J., & Young, J. B. (2004).

Equivalence of the bioimpedance and thermodilution methodsin measuring cardiac output in hospitalized patients withadvanced, decompensated chronic heart failure. AmericanJournal of Critical Care, 13, 469Y479.

Barnes, V. A., Johnson, M. H., & Treiber, F. A. (2004). Temporalstability of twenty-four-hour ambulatory hemodynamic bio-impedance measures in African American adolescents. BloodPressure Monitoring, 9, 173Y177.

Bernardy, N. C., King, A. C., & Lovallo, W. R. (2003).Cardiovascular responses to physical and psychological stressin female alcoholics with transitory hypertension after earlyabstinence. Alcoholism: Clinical and Experimental Research,27, 1489Y1498.

Bishop, G. D., Pek, J., & Ngau, F. (2005). Ethnicity, sex, traitanger, and nocturnal blood pressure decline. Psychophysiology,42, 290Y297.

Blumenthal, J. A., Jiang, W., Waugh, R. A., Frid, D. J., Morris,J. J., Coleman, R. E., et al. (1995). Mental stress-inducedischemia and ambulatory ischemia during daily life: Associationand hemodynamic features. Circulation, 92, 2102Y2108.

Boomsma, D. I., Beem, A. L., van den Berg, M., Dolan, C. V.,Koopmans, J. R., Vink, J. M., et al. (2000). Netherlands twinfamily study of anxious depression (NETSAD). Twin Research,3, 323Y334.

Brown, C. V. R., Martin, M. J., Shoemaker, W. C., Wo, C. C. J.,Chan, L., Azarow, K., et al. (2005). The effect of obesity onbioimpedance cardiac index. American Journal of Surgery,189(5), 547Y551.

Brown, C. V. R., Shoemaker, W. C., Wo, C. C. J., Chan, L., &Demetriades, D. (2005). Is noninvasive hemodynamic moni-toring appropriate for the elderly critically injured patient?Journal of Trauma Injury, Infection, and Critical Care, 58,102Y107.

Cacioppo, J. T., Hawkley, L. C., Crawford, E., Ernst, J. M.,Burleson, M. H., Kowalewski, R. B. et al. (2002). Lonelinessand health: Potential mechanisms. Psychosomatic Medicine, 64,407Y417.

Carrington, M., Walsh, M., Stambas, T., Kleiman, J., & Trinder,J. (2003). The influence of sleep onset on the diurnal variationin cardiac activity and cardiac control. Journal of SleepResearch, 12, 213Y221.

Davis, M. E., Richards, A. M., Nicholls, M. G., Yandle, T. G.,Frampton, C. M., & Troughton, R. W. (2006). Introduction ofmetoprolol increases plasma B-type cardiac natriuretic peptidesin mild, stable heart failure. Circulation, 113, 977Y985.

Deville, W. L., Bezemer, P. D., & Bouter, L. M. (2000).Publications on diagnostic test evaluation in family medicinejournals: An optimal search strategy. Journal of ClinicalEpidemiology, 53, 65Y69.

Farbey, R. (1993). Searching the literature. Be creative withMedline. BMJ, 307, 66.

Gassert, C. A. (1990). Reliability and validity of physiologicmeasurement. Critical Care Nursing Quarterly, 12, 17Y20.

Greenhalgh, T. (1997). How to read a paper: The Medlinedatabase. BMJ, 315, 180Y183.

Hamer, M., Jones, J., & Boutcher, S. (2006). Acute exer-cise reduces vascular reactivity to mental challenge in off-spring of hypertensive families. Journal of Hypertension, 24,315Y320.

Hawkley, L. C., Burleson, M. H., Bernston, G. G., & Cacioppo,J. T. (2003). Loneliness in everyday life: Cardiovascular ac-tivity, psychosocial context, and health behaviors. Journal ofPersonality and Social Psychology, 85, 105Y120.

Haynes, R. B., Wilczynski, N., McKibbon, K. A., Walker, C. J., &Sinclair, J. C. (1994). Developing optimal search strategies



for detecting clinically sound studies in MEDLINE. Journal ofthe American Medical Informatics Association, 1, 447Y458.

Higgins, J. P. T., & Green, S. (2005). Cochrane handbook forsystematic reviews of interventions 4.2.4. The CochraneLibrary 2. Chichester, UK: Wiley.

Hill, M. N. (1988). Evaluating quality of measuring tools:Reliability and validity. Cardiovascular Nursing, 24, 47Y48.

Kelsey, R. M., Alpert, B. S., Patterson, S. M., & Barnard, M.(2000). Racial differences in hemodynamic responses toenvironmental thermal stress among adolescents. Circulation,101, 2284Y2289.

Kizakevich, P. N., McCartney, M. L., Hazucha, M. J., Sleet, L. H.,Jochem, W. J., Hackney, A. C., et al. (2000). Noninvasiveambulatory assessment of cardiac function in healthy menexposed to carbon monoxide during upper and lower bodyexercise. European Journal of Applied Physiology, 83, 7Y16.

Kubicek, W. G., Karnegis, J. N., Patterson, R. P., Witsoe, D. A.,& Mattson, R. H. (1966). Development and evaluation of animpedance cardiac output system. Aerospace Medicine, 37,1208Y1212.

Lasater, M., & Von Rueden, K. T. (2003). Outpatient cardiovas-cular management utilizing impedance cardiography. AACNClinical Issues, 14, 240Y250.

Light, K. C., & Sherwood, A. (1989). Race, borderline hyper-tension, and hemodynamic responses to behavioral stressbefore and after beta-adrenergic blockade. Health Psychology,8, 577Y595.

Lowe, H. J., & Barnett, G. O. (1994). Understanding and usingthe medical subject headings (MeSH) vocabulary to performliterature searches. JAMA, 271, 1103Y1108.

McFetridge, J., & Sherwood, A. (1999). Impedance cardiographyfor noninvasive measurement of cardiovascular hemodynamics.Nursing Research, 48, 109Y113.

McFetridge, J. A., & Sherwood, A. (2000). Hemodynamic andsympathetic responses to stress during the menstrual cycle.AACN Clinical Issues, 11, 158Y167.

Nakonezny, P. A., Kowalewski, R. B., Ernst, J. M., Hawkley, L. C.,Lozano, D. L., Litvack, D. A., et al. (2001). New ambulatoryimpedance cardiograph validated against the Minnesota Imped-ance Cardiograph. Psychophysiology, 38, 465Y473.

Obrist, P. A., Light, K. C., James, S. A., & Strogatz, D. S. (1987).Cardiovascular responses to stress: I. Measures of myocardialresponse and relationship to high resting systolic pressure andparental hypertension. Psychophysiology, 24, 65Y78.

Portney, L. G., & Watkins, M. P. (1993). Validity of measure-ments. In L. G. Portney & M. P. Watkins, Foundations ofclinical research: Applications to practice (pp. 69Y86). Norwalk,CT: Appleton & Lange.

Raymond, J., Davis, G. M., & van der Plas, M. (2002).Cardiovascular responses during submaximal electrical stimu-lation-induced leg cycling in individuals with paraplegia.Clinical Physiology and Functional Imaging, 22, 92Y98.

Riese, H., Groot, P. F. C., van den Berg, M., Kupper, N. H. M.,Magnee, E. H. B., Rohaan, E. J., et al. (2003). Large-scale en-

semble averaging of ambulatory impedance cardiograms.Behavior Research Methods, Instruments, & Computers, 35,467Y477.

Sherwood, A., Allen, M. T., Fahrenberg, J., Kelsey, R. M.,Lovallo, W. R., & van Doornen, L. J. (1990). Methodologicalguidelines for impedance cardiography. Psychophysiology, 27,1Y23.

Sherwood, A., Hughes, J. W., & McFetridge, J. (2003). Ethnicdifferences in the hemodynamic mechanisms of ambulatoryblood pressure regulation. American Journal of Hypertension,16, 270Y273.

Sherwood, A., McFetridge, J., & Hutcheson, J. S. (1998).Ambulatory impedance cardiography: A feasibility study.Journal of Applied Physiology, 85, 2365Y2369.

Sherwood, A., & Turner, J. R. (1995). Hemodynamic responsesduring psychological stress: Implications for studying diseaseprocesses. International Journal of Behavioral Medicine, 2,193Y218.

Shoemaker, W. C., Belzberg, H., Wo, C. C., Milzman, D. P.,Pasquale, M. D., Baga, L., et al. (1998). Multicenter study of non-invasive monitoring systems as alternatives to invasive moni-toring of acutely ill emergency patients. Chest, 114, 1643Y1652.

Shoemaker, W. C., Wo, C. C., Bishop, M. H., Asensio, J.,Demetriades, D., Appel, P. L., et al. (1996). Noninvasivephysiologic monitoring of high-risk surgical patients. Archivesof Surgery, 131, 732Y737.

Shoemaker, W. C., Wo, C. C., Chan, L., Ramicone, E., Kamel,E. S., Velmahos, G. C., et al. (2001). Outcome prediction ofemergency patients by noninvasive hemodynamic monitoring.Chest, 120, 528Y537.

Straneva, P., Hinderliter, A., Wells, E., Lenahan, H., & Girdler, S.(2000). Smoking, oral contraceptives, and cardiovascularreactivity to stress. Obstetrics & Gynecology, 95, 78Y83.

The Evidence-Based Medicine Working Group. (2002). Users’guides to the medical literature: Essentials of evidence-basedclinical practice. Chicago: American Medical Association.

Van De Water, J. M., Miller, T. W., Vogel, R. L., Mount, B. E., &Dalton, M. L. (2003). Impedance cardiography: The next vitalsign technology? Chest, 123, 2028Y2033.

Velmahos, G. C., Wo, C. C., Demetriades, D., Bishop, M. H., &Shoemaker, W. C. (1998). Invasive and noninvasive hemody-namic monitoring of patients with cerebrovascular accidents.Western Journal of Medicine, 169, 17Y22.

Vinet, A., Nottin, S., Lecoq, A. M., Guenon, P., & Obert, P.(2001). Reproducibility of cardiac output measurements byDoppler echocardiography in prepubertal children and adults.International Journal of Sports Medicine, 22, 437Y441.

Vrijkotte, T. G. M., van Doornen, L. J. P., & de Geus, E. J. C.(2000). Effects of work stress on ambulatory blood pressure,heart rate, and heart rate variability. Hypertension, 35, 880Y886.

Vrijkotte, T. G. M., van Doornen, L. J. P., & de Geus, E. J. C.(2004). Overcommitment to work is associated with changes incardiac sympathetic regulation. Psychosomatic Medicine, 66,656Y663.



nursing research july/august 2006 vol 55, no 4, 283–291 · hemodynamics relatively unobtrusively...

Documents