The American Journal of Surgery (2011) 201, 592–598

North Pacific Surgical Association

Validation of noninvasive hemoglobin measurements

using the Masimo Radical-7 SpHb Station
Marlin Wayne Causey, M.D.a,*, Seth Miller, M.D.a, Andrew Foster, M.D.b,
Alec Beekley, M.D.a, David Zenger, M.D.b, Matthew Martin, M.D.a
a
Department of Surgery, Madigan Army Medical Center, Building 9040, Fitzsimmons Drive, Tacoma, WA 98431,
USA; bDepartment of Anesthesia, Madigan Army Medical Center, Tacoma, WA, USA

KEYWORDS: Abstract
Hemoglobin; BACKGROUND: Hemoglobin levels must be obtained through blood draws, which are invasive,
Noninvasive monitor; time-consuming, and provide only 1 data point at a time rather than continuous measurements. The
Masimo Radical-7; Masimo Radical-7 SpHb Station (Masimo Corporation, Irvine, CA) has been shown by its manufac-
Hemorrhage turers to provide accurate noninvasive hemoglobin measurements in physiologically normal patients.
The objective of this study was to validate noninvasive hemoglobin measurements using the Masimo
Radical-7 device.
METHODS: Data were prospectively collected in 2 cohorts of patients: major operations requiring
hemodynamic monitoring (operating room [OR]) and critically ill patients (intensive care unit [ICU]).
Noninvasive hemoglobin measurements (SpHb) were recorded and were then compared with laboratory
hemoglobin measurements.
RESULTS: Data were collected on 60 patients (OR ⫽ 25 and ICU ⫽ 45). The overall correlation of
the Masimo SpHb and the laboratory Hb was .77 (P ⬍ .001) in the OR group with a mean difference
of .29 g/dL (95% confidence interval [CI], .08 –.49). The overall correlation in the ICU group was .67
(P ⬍ .001) with a mean difference of .05 g/dL (95% CI, ⫺.22 to ⫺.31).
CONCLUSIONS: Noninvasive hemoglobin monitoring is a new technology that correlated with
laboratory values and supports the continued study of noninvasive hemoglobin monitoring.
© 2011 Elsevier Inc. All rights reserved.

Total hemoglobin (SpHb) is one of the most frequently
ordered laboratory tests. Currently, hemoglobin levels must
be obtained through blood draws that are invasive and
time-consuming and can only provide data at 1 time point
rather than a continuous measurement. Since the introduc-
tion of pulse oximetry, continuous noninvasive monitoring
of blood oxygenation in the operating room has become a
Presented at the 97th Annual Meeting of the North Pacific Surgical
Association, November 12–13, 2010, Tacoma, WA.
* Corresponding author. Tel.: ⫹1-253-968-2200; fax: ⫹1-253-968-
5337.
E-mail address: mwcausey@msn.com
Manuscript received October 11, 2010; revised manuscript January 14,
2011
standard of care.1 Throughout the past 2 decades, several
attempts have been made to find a suitable alternative to
laboratory hemoglobin as well as provide a continuous
noninvasive hemoglobin monitor for use in clinical medi-
cine.2–9 A noninvasive hemoglobin monitor has the po-
tential to provide continuous, noninvasive hemoglobin
measurements that would provide immediate clinical infor-
mation prompting more rapid medical intervention.
The Radical-7 Pulse CO-Oximeter (Masimo Corpora-
tion, Irvine, CA; herein referred to as “device”) uses signal
extraction technology pulse oximetry invented in 1989,
which enables the use of adaptive filters to isolate arter-
ial signals using parallel processing engine technology.
Masimo Rainbow technology, which uses multiple (7⫹)

0002-9610/$ - see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.amjsurg.2011.01.020
M.W. Causey et al. Masimo Radical-7 SpHb Station
Figure 1 Sample collection of patient data every second throughout a total hip arthroplasty with 800ml blood loss and 200ml Cell Saver®
bolus transfusion. Red line denotes Masimo Radical-7 measurements and blue triangles represent measured laboratory hemoglobin.
wavelengths of light, enables the device to noninvasively
measure SpHb, carboxyhemoglobin (SpCO), and methemo-
globin (SpMet) in addition to oxyhemoglobin and pulse
rate.10 The use of this technology to determine and follow
SpHb continuously is of clinical interest in all areas of
medicine but particularly in the operating room in patients
undergoing major surgical procedures.
The device has been previously shown to provide vali-
dated hemoglobin values in physiologically normal patients.
However, there are little data available on the accuracy and
reliability of this device in patients with severe illness or
undergoing major operative interventions. The objective of
this study was to analyze the agreement and correlation
between the SpHb measurement and the gold standard lab-
oratory hemoglobin measurement in a series of patients
undergoing major surgical procedures.
Methods
Data for this study were collected on elective operating
room cases and in the intensive care unit (ICU) as part of a
quality-assurance process. Inclusion criteria for this process
included patients who were scheduled to undergo an elec-
tive major surgical procedure under general anesthesia re-
quiring an arterial catheter and patients in the ICU with
arterial catheters. A Rainbow R 1 25 adult adhesive sensor
(Rev. C) connected to a Radical-7 Pulse CO-Oximeter (MX
board version 7.5.0.3) was placed on the middle finger of
the nondominant hand and covered with an opaque probe
cover to eliminate interference from ambient light for the
operating room. In the ICU, 3 separate probes were used
initially: the Rainbow dc-3 SC-360 (reusable sensor), then
593
the Rainbow R 1 25 adult adhesive sensor (original single-
use sensor), and finally the upgraded Rainbow 1 25 adult
adhesive sensor (updated single-use sensor). The recorded
data were then collected every second on a universal serial
bus–linked computer system using data-collection software
(ADC, Masimo Corporation) and transferred into a Mi-
crosoft Excel file (Microsoft Office; Microsoft Corporation,
Seattle, WA) for graphing and analysis (Fig. 1). During all
the operating room cases, blood collection was performed
hourly from the radial artery catheter. In the ICU, the blood
collection was performed based on clinical necessity, typi-
cally every 4 to 12 hours. Blood samples were sent imme-
diately to the hospital central laboratory for the determina-
tion of SpHb. A small sample of patients in the operating
room also had one or more hemoglobin measurements ob-
tained using the I-stat (Abbott; Abbott Park, IL) point-of-
care blood analyzer. Intraoperatively, the presence or ab-
sence of “significant bleeding,” which was defined as blood
loss greater than 200 mL, was noted for each patient.
Routine demographic, admitting, and procedure-specific
data were collected for each subject. Statistical analysis was
performed to determine the relationship between the non-
invasive hemoglobin measurement and the standard labora-
tory measurement. Given the differing sizes of the sample
populations, correlation was analyzed using the Spearman
rank correlation for the operating room and Pearson corre-
lation for the ICU. To avoid autocorrelation within the
operating room population, correlations were calculated for
hourly time intervals with an overall correlation determined
by Fisher z-transformation. Within all groups, partial cor-
relations were also performed to determine the impact of
demographic and physiological variables. A paired Student
t test was used to evaluate the agreement (mean of the
594 The American Journal of Surgery, Vol 201, No 5, May 2011

Table 1 Surgery service demographics

ysis was performed with PASW Version 18.0 (SPSS, Inc,
Chicago, IL) with statistical significance set at an alpha
Surgery service Percentage of cases level of .05.
General surgery 28
Urology 20
Vascular surgery 24
Neurosurgery 12 Results
Orthopedic surgery 8
Otolaryngology 8 Over the 3-month study period, data from 70 patients were
collected (260 data points) and divided into 25 elective oper-
ating room cases (101 laboratory data points) and 45 critical
care patients (159 data points). Fifty-five percent of the patients
differences) between SpHb and laboratory hemoglobin were men with a mean age of 64 ⫾ 18 years (range 18 –92).
measurements. Graphic representation was performed using Within the operating room group, 1 patient was American
the Bland-Altman method of bias and concordance limits Society of Anesthesiologists (ASA) class 1, 14 patients were
between the laboratory hemoglobin and device to assess for ASA class 2, 7 patients were ASA class 3, and 2 patients ASA
agreement between the 2 measurements. All statistical anal- class 4. Table 1 shows the breakdown of procedures by the

Figure 2 Correlation values of the different patient groups and the type of sensor used. Middle line is linear correlation and two outer
lines are the 95% confidence interval. The overall correlation of the Masimo Radical-7 SpHb measurement and the laboratory evaluation
in the was .78 (p ⬍ .001). (A) The overall correlation with the reusable sensor did not change significantly when adjusted for age, gender,
and the clinical parameters temperature, heart rate, or blood pressure (Temp/HR/BP). (B) The original single use sensor did not change with
age or gender but the correlation increased when controlled for Temp/HR/BP. (C) The upgraded single use sensor did not change
significantly when controlled for age, gender, or Temp/HR/BP. (D) The overall correlation of the Masimo Radical-7 SpHb measurement
and the laboratory evaluation in the operating room population did not change significantly either with age, gender, or Temp/HR.
M.W. Causey et al. Masimo Radical-7 SpHb Station 595

Table 2 Overall and service specific correlation and difference

Difference
Variables Comparison Mean n (95% confidence interval) Significance Correlation Significance
Overall Laboratory Hb 11.83 101
Masimo Hb 11.54 101 .29 (.09–.50) .01 .77 ⬍.001
Masimo vs iStat Masimo Hb 11.27 27
iStat Hb 11.25 27 .02 (⫺.28 to .33) .87 .86 ⬍.001
Laboratory vs iStat Laboratory Hb 11.34 27
iStat Hb 11.25 27 .09 (⫺.17 to .34) .48 .85 ⬍.001
General surgery Laboratory Hb 11.40 25
Masimo Hb 10.60 25 .80 (.4–1.21) .00 .84 ⬍.001
Urology Laboratory Hb 12.67 23
Masimo Hb 12.49 23 .18 (⫺.02 to .07) .07 .96 ⬍.001
Vascular surgery Laboratory Hb 11.87 23
Masimo Hb 11.27 23 .60 (⫺.05 to 1.25) .07 .44 .037
Otolaryngology Laboratory Hb 10.60 12
Masimo Hb 10.67 12 ⫺.73 (⫺.33 to .18) .542 .88 ⬍.001
Neurosurgery Laboratory Hb 12.28 10
Masimo Hb 12.60 10 ⫺.32 (⫺.86 to .22) .212 .94 ⬍.001

primary surgical service. Significant bleeding was identified in
nearly half of the patients in the operating room group (44%,
11 patients), and 5 patients required an intraoperative blood
transfusion. The overall correlation of the device measurement
and the laboratory evaluation was .78 (P ⬍ .001), with a mean
difference of .15 (95% confidence interval, ⫺.03 to .32; P ⫽
.10). Low P values on overall correlation represent strong
correlation, and low P values related to difference signify the
actual measurement differences.
The overall correlation of the device measurement and
the laboratory evaluation in the operating room population
was .77 (P ⬍ .001) (Fig. 2). Autocorrelation was avoided by
combining the following time interval correlations: preop-
erative (.84, P ⬍ .001), 1 hour (.69, P ⬍ .001), 2 hours (.69,
P ⬍ .001), 3 hours (.63, P ⫽ .01), and 4 hours (.73, P ⫽
.01). The device reported measurements were on average
.29 g/dL (95% confidence interval, .08 –.49; P ⫽ .006)
lower than the standardized laboratory evaluation (Table 2).
When compared with the iStat point-of-care hemoglobin
measurement, the central laboratory hemoglobin had a cor-
relation of .86 (P ⬍ .001) and measured .87 g/dL higher
(95% confidence interval, ⫺.11 to .34; P ⫽ .48), whereas
the device SpHb had a correlation of .85 (P ⬍ .001) and
measured .02 g/dL higher (95% confidence interval, ⫺.27 to
.32; P ⫽ .86). There was no statistically significant differ-
ence between the groups (Table 2). To assess for agreement
between the laboratory and hemoglobin measurements, a
Bland-Altman plot was constructed. Using this method,
92.1% of the measurements fell within 2 standard devia-
tions of the mean difference (Fig. 3).
Within each surgical specialty (Table 1), the correlations
and difference measurements were then analyzed. Correla-
tions were calculated only for the services that had 10 or
more comparison laboratory values. The correlation for
general surgery was .84 (P ⬍ .001), urology was .96 (P ⬍
.001), otolaryngology was .88 (P ⬍ .001), neurosurgery was
.94 (P ⬍ .001), and vascular surgery was .44 (P ⬍ .001).
Measurement differences between the laboratory hemoglo-
bin and the device hemoglobin are listed in Table 2 and
measured within .5 g/dL of the laboratory except during
general surgery (difference of .8 g/dL) and vascular surgery
(difference of .6 g/dL), both with laboratory hemoglobin
above the SpHb measurement. When the data from patients
who sustained significant intraoperative hemorrhage were
analyzed, there was a significant correlation of .86 (P ⬍
.001, Fig. 4) between the SpHb and the laboratory hemo-
globin, with the laboratory hemoglobin measuring .47 g/dL
higher on average than the device SpHb (Table 3). Addi-
tionally, in patients who required an intraoperative blood
transfusion, there was a very strong correlation of .91 (P ⬍
.001) between measurements. The laboratory measurements
were .10 g/dL higher on average than the device noninva-
sive hemoglobin monitory (Table 3).
In those patients monitored within the ICU, the overall
correlation was .67 (P ⬍ .001) with a mean difference of .05
g/dL (95% confidence interval, ⫺.21 to .30; P ⫽ .73). The
patient groups were then stratified based on the type of sensor
used. The correlation value for the reusable sensor (n ⫽ 47)
was .49 (P ⫽ .001) with a mean difference of .70 g/dL (95%
confidence interval, ⫺.09 to 1.3; P ⫽ .026). For the original
single-use sensor (n ⫽ 92), the correlation value was .61 (P ⬍
.001) with a mean difference of ⫺.24 (95% confidence inter-
val, ⫺.52 to .04, P ⫽ .09), and for the updated single-use
sensor (n ⫽ 20), the correlation value was .83 (P ⬍ .001) with
a mean difference of ⫺.25 (95% confidence interval, ⫺.92 to
.41; P ⫽ .44). As with the operating room population, the
Bland-Altman plot showed that 96% of the measurements fell
within 2 standard deviations of the mean difference with the
reusable sensor, 95% with the original single-use sensor, and
95% with the upgraded single-use sensor (Fig. 3). To assess the
correlation given the impact of demographic factors and clin-
ical data, partial correlations were determined based on the
setting and in the critical care setting based on the type of
sensor used (Fig. 2).
596 The American Journal of Surgery, Vol 201, No 5, May 2011

Figure 3 Graphic representation according to the Bland and Altman method of the bias and the concordance limits between the laboratory
and Masimo Radical-7 device. Overall there were 260 valid values and 248 measurements (95%) within the upper and lower limits. In the
ICU group, 96% of the measurements fell within two standard deviations of the mean difference with the reusable sensor (A), 95% with
the original single use sensor (B), and 95% with the upgraded single use sensor (C). In the OR group, there were 101 valid values and 93
measurements (92%) within the upper and lower limits (D).

Comments fifty-eight data pairs were collected comparing laboratory CO-

The need for noninvasive monitoring is essential in the
operating room and ICU.11,12 There are 3 groups of patients in
which noninvasive hemoglobin monitoring is especially im-
portant: high-risk critically ill patients or patients undergoing
high-risk operations, critically ill bleeding/transfused patients
or patients undergoing operations with an expected high blood
loss, and patients not expected to have significant bleeding
intraoperatively but in whom obtaining laboratory values
would be difficult or impossible if not preplanned. The device
SpHb is likely of significant benefit in all 3 groups of patients.
The accuracy of the device in measuring SpHb was first
evaluated during Food and Drug Administration approval us-
ing 59 volunteers to collect 492 data pairs comparing SpHb
with laboratory hemoglobin levels. Over a range from 8 g/dL
to 17 g/dL hemoglobin, the data pairs had a correlation value
of .90.13 Additionally, the prototype SpHb probe was tested in
19 surgery and 19 healthy volunteer patients. Four hundred
Oximeter readings with the SpHb readings. The CO-Oximeter
readings ranged from 4.4 g/dL to 15.8 g/dL and had a corre-
lation value of .89.14 Although this indicates a strong correla-
tion in physiologically normal subjects, there are no indications
for use in patients with physiological alterations or disease and
undergoing surgical procedures. This is the first study to ex-
amine this technology for assessment of measuring hemoglo-
bin and hemorrhage in the operating room and ICU.
In high-risk patients, noninvasive SpHb monitoring will
benefit the patient, intensivist, and anesthesiologist by al-
lowing continuous physiological monitoring without having
to occlude the arterial line and may even preclude the need
for a central line when the purpose is only for frequent
laboratory hemoglobin analysis. In this group of patients as
well as the other aforementioned groups of patients, benefit
will be obtained by having a real-time measure of hemo-
globin instead of having to wait for the return of laboratory
M.W. Causey et al. Masimo Radical-7 SpHb Station 597

Figure 4 Laboratory values and SpHb readings correlate in patients with significant bleeding. Middle line is linear correlation and two
dashed lines are the 95% confidence interval.

results. In the patients who have invasive devices placed
simply for laboratory analysis, the device may be able to
replace these monitoring devices in select patients. The
group of patients this would most likely benefit is the ASA
class 1 and 2 patients undergoing lengthy surgeries with
moderate blood loss. A final group of patients are those who
undergo surgical procedures in which there is a sudden
unexpected hemorrhage. The probe could be placed on the
finger and easily attached yielding results in less than 2
minutes (the estimated time based on our experience) as
well as provide continuous monitoring throughout the du-
ration of the procedure.
Based on the strong correlation between device SpHb and
laboratory hemoglobin, noninvasive hemoglobin monitoring
can be beneficial in the operating room and critical care setting
and may potentially decrease the invasiveness of both elective
surgery and critical illness while providing continuous hemo-
globin monitoring in addition to standard pulse oximetry data.
Complications from access procedures are minimal, but case
reports are frequent, and in some patients these procedures
may be foregone entirely with this technology.15–18 Two sub-
sets of patients that would benefit from the use of noninvasive
hemoglobin monitoring are the ASA class 1 and 2 patients
undergoing lengthy surgical procedures because hemodynamic
changes are less frequent and dramatic and critically ill patients
who need frequent laboratory hemoglobin monitoring without
invasive central access or arterial hemodynamic monitoring.
The results of this study are limited, but this area of medical
technology represents a significant advancement in hemoglo-
bin monitoring and technology.
The limitations of this study are the variability of the mea-
surements reported with both the laboratory hemoglobin and
the SpHb. SpHb accuracies were not validated under motion or
low perfusion states and are accurate to 1 g/dL in no-motion
environments. This limitation of the device was minimized by
being studied in a low-motion environment of the operating
room, and under these conditions, we were able to obtain
statistical significance, which was attainable in almost all the

Table 3 Correlation and difference in patients sustaining significant bleeding

Mean N* Difference (95% confidence interval) P value Correlation Significance

Bleeding group
Laboratory Hb 11.52 50
Masimo Hb 11.05 50 .47 (.207–.733) ⬍.001 .855 ⬍.001
Transfusion group
Laboratory Hb 11.13 21
Masimo Hb 11.11 21 .10 (⫺.1966 to .2271) .882 .906 ⬍.001
*N is the total number of laboratory Hb to SpHb evaluations.
598
statistical analyses secondary to the correlations between SpHb
and laboratory hemoglobin measurements. In the intensive
care setting, movement was more frequent, and hemoglobin
monitoring was maximized by using the updated single-use
sensors. Although all the correlations achieved statistical sig-
nificance across all surgery types, this was not the case for
measurement differences. Overall, there was a difference in the
values reported form the device and the laboratory measure-
ments, but only 1 service, general surgery, had enough samples
to determine the true difference in the actual measurements.
The other services had laboratory values that were close to the
actual laboratory values but not enough sample size to deter-
mine the actual difference but neared significance with uro-
logic and vascular surgeries. One of the strengths of this study
is the controlled physiological setting of the operating room
population. Having many of the potentially modifiable physi-
ological factors controlled in an operating room setting, this
allowed for the best assessment of this new technology in a
no-motion environment, for which the device received Food
and Drug Administration approval. There was also a diverse
measurement of SpHb in the surgical procedures and in the
ICU, and this allowed for measurement across multiple clinical
scenarios. Furthermore, 44% of patients in the study sustained
significant intraoperative blood loss, the patient population for
which the device would prove to be most useful.
Conclusions
Noninvasive hemoglobin monitoring is a new technology
with a good indication in the operating room and ICU for
providing immediate and continuous clinical information on
hemoglobin values allowing for more rapid and appropriate
medical intervention in the properly chosen patient population.
Our study found the device SpHb to have clinically acceptable
accuracy during most types of surgery including those with
significant blood loss. The full understanding of the utility of
the device is not fully known, but this study supports the
continued study of continuous and noninvasive hemoglobin
monitoring systems.
References
1. Carlson KA, Jahr JS. A historical overview and update on pulse
oximetry. Anesthesiol Rev 1993;20:173– 81.
2. McMurdy JW, Jay GD, Suner S, et al. Noninvasive optical, electrical,
and acoustic methods of total hemoglobin determination. Clin Chem
2008;54:264 –72.
3. Noiri E, Kobayashi N, Takamura Y, et al. Pulse total-hemoglobinom-
eter provides accurate noninvasive monitoring. Crit Care Med 2005;
33:2831–5.
4. Cerussi A, Van Woerkom R, Waffarn F, et al. Noninvasive monitoring
of red blood cell transfusion in very low birthweight infants using
diffuse optical spectroscopy. J Biomed Opt 2005;10:051401.
5. Saigo K, Imoto S, Hashimoto M, et al. Noninvasive monitoring of
hemoglobin. The effects of WBC counts on measurement. Am J Clin
Pathol 2004;121:51–5.
The American Journal of Surgery, Vol 201, No 5, May 2011
6. Esenaliev RO, Petrov YY, Hartrumpf O, et al. Continuous, noninva-
sive monitoring of total hemoglobin concentration by an optoacoustic
technique. Appl Opt 2004;43:3401–7.
7. Kajiume T, Kawano Y, Takaue Y, et al. Continuous monitoring of
hematocrit values during apheresis for allogeneic peripheral blood
stem cell collection. J Hematother 1998;7:493–7.
8. Edwards AD, Richardson C, van der Zee P, et al. Measurement of
hemoglobin flow and blood flow by near-infrared spectroscopy. J Appl
Physiol 1993;75:1884 –9.
9. Kasler M, von Glass W, Albrecht HP, et al. [Noninvasive intraoper-
ative measurement of intracapillary hemoglobin oxygenation and rel-
ative hemoglobin concentration in surgical skin flaps]. HNO 1990;38:
375– 8.
10. Clinical Studies, Masimo SET. Pulse oximeter bibliography. Available
at: http://www.masimo.com/cpub/clinicals.htm. Accessed January 14,
2011.
11. Shoemaker WC, Wo CC, Bishop MH, et al. Noninvasive physiologic
monitoring of high-risk surgical patients. Arch Surg 1996;131:732–7.
12. Schnapp LM, Cohen NH. Pulse oximetry. Uses and abuses. Chest
1990;98:1244 –50.
13. Masimo Technical Bulletin. Accuracy of noninvasive and continuous
hemoglobin measurements by pulse CO-Oximetry. No. 7669-5063C-
1209. Available at: http://www.masimo.com/pdf/SpHb/LAB5063C.
pdf. Accessed January 14, 2011.
14. Macknet MR, K-JP, Applegate RL, et al. Non-invasive measurement
of continuous hemoglobin concentration via pulse CO-oximetry. An-
esthesiology 2007;107:A1545.
15. Muzolf J, Onichimowski D, Podlinska I. [Preoperative risk evaluation
in cardiac patients scheduled for vascular surgery]. Anestezjol Intens
Ter 2008;40:103–7.
16. Mirbagheri N, Hoy C. Hand ischaemia post-radial artery cannulation:
a cautionary reminder of an uncommon complication of a common
procedure. ANZ J Surg 2008;78:719 –20.
17. Valentine RJ, Modrall JG, Clagett GP. Hand ischemia after radial
artery cannulation. J Am Coll Surg 2005;201:18 –22.
18. Bowdle TA. Complications of invasive monitoring. Anesthesiol Clin
North America 2002;20:571– 88.
Discussion
Amir Bastawrous, M.D. (Seattle, WA): The use of point
of care techniques to evaluate patient physiology is not new.
As a profession, physicians are impatient, and surgeons more
so than most. Furthermore, the acquisition of accurate patient
data in an expeditious manner does more than simply address
our collective impatience. Particularly in situations in which
conditions are rapidly changing, the differences in waiting time
between a hemoglobin result obtained from the laboratory and
a point-of-care iStat reading or a transcutaneous measure of
hemoglobin may be important and could significantly impact
physician intervention and patient outcome.
At the same time, consideration must be made of the addi-
tional costs incurred with the incorporation of new technolo-
gies, especially in an environment of uncertain reimbursement.
So, the question becomes does the device tested in this study
affect patient outcome enough to justify its added cost? The
future of new technologies must include a cost analysis. How-
ever, the validating results of this comparative analysis are
promising. The addition of accurate instantaneous hemoglobin
measurement may facilitate rapid patient-care decisions and
improve outcomes.