A novel method of blood pressure measurement in patients with continuous-flow left ventricular assist devices
2014; Elsevier BV; Volume: 33; Issue: 11 Linguagem: Inglês
10.1016/j.healun.2014.08.011
ISSN1557-3117
AutoresKei Woldendorp, Sunil Gupta, Jacqueline Lai, K. Dhital, Chris Hayward,
Tópico(s)Cardiac Structural Anomalies and Repair
ResumoContinuous-flow LVADs (cfLVADs) have largely replaced pulsatile pumps for chronic mechanical circulatory support in view of improved post-operative morbidity and mortality.1Slaughter M.S. Pagani F.D. Rogers J.G. et al.Clinical management of continuous-flow left ventricular assist devices in advanced heart failure.J Heart Lung Transplant. 2010; 29: S1-39Abstract Full Text Full Text PDF PubMed Scopus (715) Google Scholar However, one of the major issues associated with the reduced pulsatility of cfLVADs is non-invasive blood pressure (BP) measurement. Doppler ultrasound of the radial artery combined with a manual sphygmomanometry is the accepted non-invasive method for BP monitoring1Slaughter M.S. Pagani F.D. Rogers J.G. et al.Clinical management of continuous-flow left ventricular assist devices in advanced heart failure.J Heart Lung Transplant. 2010; 29: S1-39Abstract Full Text Full Text PDF PubMed Scopus (715) Google Scholar but requires training and dexterity for accurate results. Because of its simplicity and widespread availability, we propose using digital pulse oximetry combined with manual sphygmomanometry to estimate BP in cfLVAD patients.Thirty patients implanted with HeartWare HVAD (HeartWare International Inc, Framingham, MA) pumps were studied in the intensive care unit (ICU) and in the outpatient setting (Supplementary Material, Table 1, available on the jhltonline.org Web site). The study was approved by the St. Vincent's and Mater Health Human Research Ethics Committee. Fourteen patients were studied in the intensive care unit (ICU). Each patient had intra-arterial pressure, pulse oximetry, Doppler ultrasound, and automated oscillometric recorded in that order. Eight of the ICU patients, as well as 16 further patients, underwent non-invasive reproducibility studies as outpatients (Figure 1). Devices used are reported in the Supplementary Material (available on the jhltonline.org Web site). Three recordings were attempted for each device by 2 individuals. If a manual cuff was necessary for any of the recordings (pulse oximetry and Doppler), the cuff was evenly deflated at a slow rate of between 2 and 4 mm Hg per second. Aortic valve status was obtained from the most recent available echocardiogram.Repeatability studies were performed and analyzed using one-way analysis of variance. The data for discrete variables were divided into 2 groups and analyzed with the Mann-Whitney test for unpaired data. Prism 6.0a software (GraphPad Software, Inc, La Jolla, CA) was used for all statistical analysis.ICU recordings were taken a median of 21 days after LVAD implant, with 11 patients studied during the initial implant (median, 8 days; range, 1–50 days) and 3 studied during repeat ICU admission (median, 148 days; range 90–182 days) after LVAD implant.The mean arterial pressure (MAP) by each of the recording methods was similar (Figure 2a). The automated oscillometric systolic blood pressure (Auto SBP) was slightly higher than the intra-arterial pressure (p = 0.038); however, there were no other significant differences (p = 0.056; Figure 2b). MAP by the pulse oximetry method correlated well with intra-arterial MAP (R2 = 0.41, p = 0.0016; Figure 3a). The pulse oximetry MAP was slightly higher than the intra-arterial MAP on Bland-Altman analysis (3.2 ± 7.9 mm Hg; 95% confidence interval, –12.3 to 18.6 mm Hg; Figure 3a). Doppler also correlated well with intra-arterial MAP (R2 = 0.40, p = 0.029; Figure 3b) and was marginally higher than intra-arterial MAP on Bland-Altman analysis (4.0 ± 9.0 mm Hg; 95% confidence interval [CI], –13.6 to 21.6 mm Hg; Figure 3b). Pulse oximetry demonstrated excellent correlation with Doppler (R2 = 0.91, p = 0.0001; Figure 4a), and Bland-Altman analysis also revealed a similar relationship, with almost identical overall recordings (–0.3 ± 3.6 mm Hg; 95% CI, –7.3 to 6.7 mm Hg). Auto SBP readings correlated poorly with intra-arterial MAP readings (R2 = 0.19, p = 0.33) or Auto SBP with intra-arterial SBP (R2 = 0.17, p = 0.35). Auto BP in the ICU was unreliable, with only 20 successful readings in 63 attempts (32%).Figure 2Mean blood pressure (BP) between different methods compared against (a) mean (MAP) and (b) systolic arterial blood pressure (SBP). ART, intra-arterial pressure; aSBP, Intra-arterial systolic blood pressure; AUTOMAP, automated oscillometric cuff mean arterial pressure; AUTOSBP, automated oscillometric cuff systolic arterial pressure; DOPPLER, Doppler sphygmomanometry; PULSE, pulse oximetry with sphygmomanometry.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 3Comparison of linear regression (left) and Bland-Altman analysis (right) between (a) intra-arterial (ART) mean arterial pressure and pulse oximetry (PULSE) recordings and (b) between ART mean arterial pressure and pulsed Doppler (DOPPLER) estimates. The dashed lines indicate the 95% confidence intervals.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 4Comparison of linear regressions between pulse oximetry (PULSE) and Doppler (DOPPLER) recordings in the (a) intensive care unit and (b) outpatient settings. Multiple recordings were taken in some patients. The dashed lines indicate the 95% confidence intervals.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Non-invasive BP measurements were taken in 24 outpatients at a median of 117 days (range, 8–1,074 days). There was a strong correlation between Doppler and pulse oximetry (R2 = 0.90, p = 0.0001; Figure 4b), with very similar results on Bland-Altman analysis (bias, –0.8 ± 2.7 mm Hg; 95% CI, –6.1 to 4.4 mm Hg). Auto was only successful in 18 of 60 attempted recordings (30%). Results for the Auto data set are reported in the Supplementary Material (available on the jhltonline.org Web site).Repeatability of pulse oximetry technique on sets of 3 repeat measures on each patient (F2, 20 = 0.24, p = 0.76) was very good and similar to the intra-arterial recordings (F2, 20 = 0.25, p = 0.69). Repeatability was assessed twice in a series of 13 patients, with an initial recording at median of 50 days (range, 8–273 days) and a second recording at 251 days (range, 22–651 days). A strong correlation remained between Doppler and pulse oximetry BP estimates (p < 0.001 for both assessments), with mean differences between the 2 recordings of –0.7 mm Hg at baseline and –0.9 mm Hg at the repeat assessment.We have demonstrated the feasibility, accuracy, and reproducibility of the finger pulse oximetry waveform to measure BP in patients supported with cfLVADs. BP estimates were similar to the current standard Doppler ultrasound BP with brachial sphygmomanometry.1Slaughter M.S. Pagani F.D. Rogers J.G. et al.Clinical management of continuous-flow left ventricular assist devices in advanced heart failure.J Heart Lung Transplant. 2010; 29: S1-39Abstract Full Text Full Text PDF PubMed Scopus (715) Google Scholar We believe this technique warrants further consideration due to ease of use and widespread availability of oximetry devices in emergency departments and in outpatient care settings.Alternate techniques investigated for cfLVAD BP measurement include a Doppler-type device housed in a wristwatch-type strap that allows continuous location of the ultrasound signal. This device was validated in clinical trials, but limited availability has prevented widespread use.2Schima H. Boehm H. Huber L. et al.Automatic system for noninvasive blood pressure determination in rotary pump recipients.Artif Organs. 2004; 28: 451-457Crossref PubMed Scopus (15) Google Scholar The Nexfin (Edwards Lifesciences, Irvine, CA) a next-generation digital volume-clamp device shows promise in stable cfLVAD patients, and although MAP was underestimated compared with invasive BP by 6.9 ± 5.1 mm Hg in a cohort of 29 HeartMate II patients,3Martina J.R. Westerhof B.E. de Jonge N. et al.Noninvasive arterial blood pressure waveforms in patients with continuous-flow left ventricular assist devices.ASAIO J. 2014; 60: 154-161Crossref PubMed Scopus (17) Google Scholar this compares favorably with Doppler, which was recently shown to be higher than the MAP by 9.5 mm Hg.4Lanier G.M. Orlanes K. Hayashi Y. et al.Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device.Circ Heart Fail. 2013; 6: 1005-1012Crossref PubMed Scopus (85) Google ScholarA recent study examining the Elemano slow-cuff deflation oscillometric device (Terumo Medical Corp, Somerset, NJ) demonstrated a good estimate of intra-arterial MAP, with successful recording in 91% of attempts.4Lanier G.M. Orlanes K. Hayashi Y. et al.Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device.Circ Heart Fail. 2013; 6: 1005-1012Crossref PubMed Scopus (85) Google Scholar We examined the reliability of standard hospital automatic oscillometric BP devices and found they failed to record BP in 68.3% of patients in the ICU and in 70% in outpatient cohorts. Our results are similar to other automated oscillometric devices, reporting approximately 50% success,5Bennett M.K. Roberts C.A. Dordunoo D. Shah A. Russell S.D. Ideal methodology to assess systemic blood pressure in patients with continuous-flow left ventricular assist devices.J Heart Lung Transplant. 2010; 29: 593-594Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar and to the comparator device in the above study, which had 63% successful readings.4Lanier G.M. Orlanes K. Hayashi Y. et al.Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device.Circ Heart Fail. 2013; 6: 1005-1012Crossref PubMed Scopus (85) Google Scholar The poor oscillometric success in the current study, particularly in the ICU cohort, is likely to be related to the significantly lower pulsatility early after cfLVAD implant. Clearly, the rate of cuff deflation is critical in the success of automated devices in cfLVAD patients.The pulsatile plethysmographic waveform can be difficult to determine in patients with very high heart rates (>100 beats/min), those in constant atrial fibrillation, or patients with biventricular assist devices. In these patients, a slower deflation of the manual cuff (~1 to 2 mm Hg per second) was found to remedy the issue in most cases. Only in 2 patients were we unable to obtain a sufficient waveform; these 2 patients had a very high heart rate (>117 beats/min) and a biventricular assist device, respectively. On subsequent occasions, a suitable pulse oximetry reading was available from both individuals.Study conclusions are limited by sample size and the use of different machines for automated oscillometric measurements. However, we found similar failure to record rates between both oscillometric machines used. The Terumo Elemano device is not available in Australia and could not be evaluated. A further limitation is that aortic valve opening status was not available at the time of BP comparison, and pulsatility index is not reported for the HeartWare HVAD.In summary, non-invasive BP results obtained using sphygmomanometry combined with pulse oximetry corresponded well with intra-arterial measurements and were similar to sphygmomanometry combined with Doppler in estimating MAP. The widespread availability of pulse oximetry throughout emergency departments and ICUs suggests that it may be transferred into routine cfLVAD clinical care, including in non-implanting centers that may not be familiar with the use of Doppler for BP measurement. Despite the simplicity of the suggested technique, we recommend that LVAD patient management should always be coordinated through a designated implanting center.Disclosure statementThe assistance and contributions of Desiree Robson, LVAD Coordinator, in data acquisition are noted. The authors are also grateful for the assistance of Patrick Markey, Peter Macdonald, Anne Keogh, Eugene Kotlyar, Emily Granger and Phillip Spratt in the care of these patients and for allowing the study to be performed.C.S.H. has received research funding and travel support unrelated to the current project from HeartWare, Inc. None of the other authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose. Continuous-flow LVADs (cfLVADs) have largely replaced pulsatile pumps for chronic mechanical circulatory support in view of improved post-operative morbidity and mortality.1Slaughter M.S. Pagani F.D. Rogers J.G. et al.Clinical management of continuous-flow left ventricular assist devices in advanced heart failure.J Heart Lung Transplant. 2010; 29: S1-39Abstract Full Text Full Text PDF PubMed Scopus (715) Google Scholar However, one of the major issues associated with the reduced pulsatility of cfLVADs is non-invasive blood pressure (BP) measurement. Doppler ultrasound of the radial artery combined with a manual sphygmomanometry is the accepted non-invasive method for BP monitoring1Slaughter M.S. Pagani F.D. Rogers J.G. et al.Clinical management of continuous-flow left ventricular assist devices in advanced heart failure.J Heart Lung Transplant. 2010; 29: S1-39Abstract Full Text Full Text PDF PubMed Scopus (715) Google Scholar but requires training and dexterity for accurate results. Because of its simplicity and widespread availability, we propose using digital pulse oximetry combined with manual sphygmomanometry to estimate BP in cfLVAD patients. Thirty patients implanted with HeartWare HVAD (HeartWare International Inc, Framingham, MA) pumps were studied in the intensive care unit (ICU) and in the outpatient setting (Supplementary Material, Table 1, available on the jhltonline.org Web site). The study was approved by the St. Vincent's and Mater Health Human Research Ethics Committee. Fourteen patients were studied in the intensive care unit (ICU). Each patient had intra-arterial pressure, pulse oximetry, Doppler ultrasound, and automated oscillometric recorded in that order. Eight of the ICU patients, as well as 16 further patients, underwent non-invasive reproducibility studies as outpatients (Figure 1). Devices used are reported in the Supplementary Material (available on the jhltonline.org Web site). Three recordings were attempted for each device by 2 individuals. If a manual cuff was necessary for any of the recordings (pulse oximetry and Doppler), the cuff was evenly deflated at a slow rate of between 2 and 4 mm Hg per second. Aortic valve status was obtained from the most recent available echocardiogram. Repeatability studies were performed and analyzed using one-way analysis of variance. The data for discrete variables were divided into 2 groups and analyzed with the Mann-Whitney test for unpaired data. Prism 6.0a software (GraphPad Software, Inc, La Jolla, CA) was used for all statistical analysis. ICU recordings were taken a median of 21 days after LVAD implant, with 11 patients studied during the initial implant (median, 8 days; range, 1–50 days) and 3 studied during repeat ICU admission (median, 148 days; range 90–182 days) after LVAD implant. The mean arterial pressure (MAP) by each of the recording methods was similar (Figure 2a). The automated oscillometric systolic blood pressure (Auto SBP) was slightly higher than the intra-arterial pressure (p = 0.038); however, there were no other significant differences (p = 0.056; Figure 2b). MAP by the pulse oximetry method correlated well with intra-arterial MAP (R2 = 0.41, p = 0.0016; Figure 3a). The pulse oximetry MAP was slightly higher than the intra-arterial MAP on Bland-Altman analysis (3.2 ± 7.9 mm Hg; 95% confidence interval, –12.3 to 18.6 mm Hg; Figure 3a). Doppler also correlated well with intra-arterial MAP (R2 = 0.40, p = 0.029; Figure 3b) and was marginally higher than intra-arterial MAP on Bland-Altman analysis (4.0 ± 9.0 mm Hg; 95% confidence interval [CI], –13.6 to 21.6 mm Hg; Figure 3b). Pulse oximetry demonstrated excellent correlation with Doppler (R2 = 0.91, p = 0.0001; Figure 4a), and Bland-Altman analysis also revealed a similar relationship, with almost identical overall recordings (–0.3 ± 3.6 mm Hg; 95% CI, –7.3 to 6.7 mm Hg). Auto SBP readings correlated poorly with intra-arterial MAP readings (R2 = 0.19, p = 0.33) or Auto SBP with intra-arterial SBP (R2 = 0.17, p = 0.35). Auto BP in the ICU was unreliable, with only 20 successful readings in 63 attempts (32%). Non-invasive BP measurements were taken in 24 outpatients at a median of 117 days (range, 8–1,074 days). There was a strong correlation between Doppler and pulse oximetry (R2 = 0.90, p = 0.0001; Figure 4b), with very similar results on Bland-Altman analysis (bias, –0.8 ± 2.7 mm Hg; 95% CI, –6.1 to 4.4 mm Hg). Auto was only successful in 18 of 60 attempted recordings (30%). Results for the Auto data set are reported in the Supplementary Material (available on the jhltonline.org Web site). Repeatability of pulse oximetry technique on sets of 3 repeat measures on each patient (F2, 20 = 0.24, p = 0.76) was very good and similar to the intra-arterial recordings (F2, 20 = 0.25, p = 0.69). Repeatability was assessed twice in a series of 13 patients, with an initial recording at median of 50 days (range, 8–273 days) and a second recording at 251 days (range, 22–651 days). A strong correlation remained between Doppler and pulse oximetry BP estimates (p < 0.001 for both assessments), with mean differences between the 2 recordings of –0.7 mm Hg at baseline and –0.9 mm Hg at the repeat assessment. We have demonstrated the feasibility, accuracy, and reproducibility of the finger pulse oximetry waveform to measure BP in patients supported with cfLVADs. BP estimates were similar to the current standard Doppler ultrasound BP with brachial sphygmomanometry.1Slaughter M.S. Pagani F.D. Rogers J.G. et al.Clinical management of continuous-flow left ventricular assist devices in advanced heart failure.J Heart Lung Transplant. 2010; 29: S1-39Abstract Full Text Full Text PDF PubMed Scopus (715) Google Scholar We believe this technique warrants further consideration due to ease of use and widespread availability of oximetry devices in emergency departments and in outpatient care settings. Alternate techniques investigated for cfLVAD BP measurement include a Doppler-type device housed in a wristwatch-type strap that allows continuous location of the ultrasound signal. This device was validated in clinical trials, but limited availability has prevented widespread use.2Schima H. Boehm H. Huber L. et al.Automatic system for noninvasive blood pressure determination in rotary pump recipients.Artif Organs. 2004; 28: 451-457Crossref PubMed Scopus (15) Google Scholar The Nexfin (Edwards Lifesciences, Irvine, CA) a next-generation digital volume-clamp device shows promise in stable cfLVAD patients, and although MAP was underestimated compared with invasive BP by 6.9 ± 5.1 mm Hg in a cohort of 29 HeartMate II patients,3Martina J.R. Westerhof B.E. de Jonge N. et al.Noninvasive arterial blood pressure waveforms in patients with continuous-flow left ventricular assist devices.ASAIO J. 2014; 60: 154-161Crossref PubMed Scopus (17) Google Scholar this compares favorably with Doppler, which was recently shown to be higher than the MAP by 9.5 mm Hg.4Lanier G.M. Orlanes K. Hayashi Y. et al.Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device.Circ Heart Fail. 2013; 6: 1005-1012Crossref PubMed Scopus (85) Google Scholar A recent study examining the Elemano slow-cuff deflation oscillometric device (Terumo Medical Corp, Somerset, NJ) demonstrated a good estimate of intra-arterial MAP, with successful recording in 91% of attempts.4Lanier G.M. Orlanes K. Hayashi Y. et al.Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device.Circ Heart Fail. 2013; 6: 1005-1012Crossref PubMed Scopus (85) Google Scholar We examined the reliability of standard hospital automatic oscillometric BP devices and found they failed to record BP in 68.3% of patients in the ICU and in 70% in outpatient cohorts. Our results are similar to other automated oscillometric devices, reporting approximately 50% success,5Bennett M.K. Roberts C.A. Dordunoo D. Shah A. Russell S.D. Ideal methodology to assess systemic blood pressure in patients with continuous-flow left ventricular assist devices.J Heart Lung Transplant. 2010; 29: 593-594Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar and to the comparator device in the above study, which had 63% successful readings.4Lanier G.M. Orlanes K. Hayashi Y. et al.Validity and reliability of a novel slow cuff-deflation system for noninvasive blood pressure monitoring in patients with continuous-flow left ventricular assist device.Circ Heart Fail. 2013; 6: 1005-1012Crossref PubMed Scopus (85) Google Scholar The poor oscillometric success in the current study, particularly in the ICU cohort, is likely to be related to the significantly lower pulsatility early after cfLVAD implant. Clearly, the rate of cuff deflation is critical in the success of automated devices in cfLVAD patients. The pulsatile plethysmographic waveform can be difficult to determine in patients with very high heart rates (>100 beats/min), those in constant atrial fibrillation, or patients with biventricular assist devices. In these patients, a slower deflation of the manual cuff (~1 to 2 mm Hg per second) was found to remedy the issue in most cases. Only in 2 patients were we unable to obtain a sufficient waveform; these 2 patients had a very high heart rate (>117 beats/min) and a biventricular assist device, respectively. On subsequent occasions, a suitable pulse oximetry reading was available from both individuals. Study conclusions are limited by sample size and the use of different machines for automated oscillometric measurements. However, we found similar failure to record rates between both oscillometric machines used. The Terumo Elemano device is not available in Australia and could not be evaluated. A further limitation is that aortic valve opening status was not available at the time of BP comparison, and pulsatility index is not reported for the HeartWare HVAD. In summary, non-invasive BP results obtained using sphygmomanometry combined with pulse oximetry corresponded well with intra-arterial measurements and were similar to sphygmomanometry combined with Doppler in estimating MAP. The widespread availability of pulse oximetry throughout emergency departments and ICUs suggests that it may be transferred into routine cfLVAD clinical care, including in non-implanting centers that may not be familiar with the use of Doppler for BP measurement. Despite the simplicity of the suggested technique, we recommend that LVAD patient management should always be coordinated through a designated implanting center. Disclosure statementThe assistance and contributions of Desiree Robson, LVAD Coordinator, in data acquisition are noted. The authors are also grateful for the assistance of Patrick Markey, Peter Macdonald, Anne Keogh, Eugene Kotlyar, Emily Granger and Phillip Spratt in the care of these patients and for allowing the study to be performed.C.S.H. has received research funding and travel support unrelated to the current project from HeartWare, Inc. None of the other authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose. The assistance and contributions of Desiree Robson, LVAD Coordinator, in data acquisition are noted. The authors are also grateful for the assistance of Patrick Markey, Peter Macdonald, Anne Keogh, Eugene Kotlyar, Emily Granger and Phillip Spratt in the care of these patients and for allowing the study to be performed. C.S.H. has received research funding and travel support unrelated to the current project from HeartWare, Inc. None of the other authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose. Appendix A. Supplementary Materials Download .pdf (.42 MB) Help with pdf files Supplementary Material Download .pdf (.42 MB) Help with pdf files Supplementary Material
Referência(s)