Continuous Finger-cuff versus Intermittent Oscillometric Arterial Pressure Monitoring and Hypotension during Induction of Anesthesia and Noncardiac Surgery: The DETECT Randomized Trial

We conducted a single-center randomized trial in patients having noncardiac surgery with general anesthesia between March 26, 2021, and October 6, 2021, at the University Medical Center Hamburg–Eppendorf in Hamburg, Germany. The trial was approved by the ethics committee (Ethikkommission der Ärztekammer Hamburg, Hamburg, Germany, registration number PV7361), and all patients provided written informed consent. Eligible patients were approached and enrolled at least 1 day before surgery by dedicated trial personnel.

The trial was registered at ClinicalTrials.gov (NCT04736862) on February 3, 2021 (principal investigator, Karim Kouz). The statistical analysis plan was written and approved by the investigators and trial statistician before analyses began but was not publicly available. We report the trial according to the Consolidated Standards of Reporting Trials (CONSORT) statement.¹⁰ 

We included adults at least 45 yr old who were scheduled for elective noncardiac surgery with general anesthesia and in whom intermittent arterial pressure monitoring using upper-arm cuff oscillometry was planned. We chose to only include patients at least 45 yr old because baseline risk for perioperative hypotension and cardiovascular complications increases with age and because this threshold has been used in previous large trials.^11,12  We did not include patients in whom invasive arterial pressure monitoring with an arterial catheter was planned nor those in whom systolic arterial pressure differed by more than 20 mmHg between the right and left arms. Other exclusion criteria were emergency surgery, pregnancy, and heart rhythms other than sinus rhythm.

In all patients, we initiated both continuous finger-cuff and intermittent oscillometric arterial pressure monitoring before induction of anesthesia. Patients were then randomized to unblinded continuous finger-cuff arterial pressure monitoring or to intermittent oscillometric arterial pressure monitoring (with blinded continuous finger-cuff arterial pressure monitoring) in a 1:1 ratio without stratification based on computer-generated codes. Assignments were concealed in sequentially numbered opaque envelopes that were prepared by personnel not involved in the trial. The envelopes were opened shortly before induction of anesthesia by dedicated trial personnel, who were otherwise not involved in the care of the patients, thus concealing allocation as long as practical.

In patients assigned to continuous finger-cuff monitoring, arterial pressure waveforms and measurements from the finger-cuff were displayed on the patient monitor (Infinity Delta; Dräger Medical, Germany), and the treating anesthesiologist was blinded to intermittent oscillometric arterial pressure monitoring. In patients assigned to intermittent oscillometric monitoring, oscillometric arterial pressure measurements were displayed on the patient monitor, and the treating anesthesiologist was blinded to continuous finger-cuff arterial pressure monitoring.

All patients had general anesthesia, and there was no anesthesia management protocol. General anesthesia was induced with propofol supplemented with sufentanil, fentanyl, or remifentanil. In patients requiring tracheal intubation, a neuromuscular blocking agent (rocuronium, mivacurium, or succinylcholine) was given. General anesthesia was maintained with a continuous propofol infusion or inhaled sevoflurane. Most patients had two peripheral venous catheters: one for fluid and nonvasoactive drug administration and one for norepinephrine administration. The oscillometric upper-arm cuff was usually placed on the arm where fluids and nonvasoactive drugs were administered. There was no specific hypotension treatment algorithm, but our institutional routine is to maintain MAP above 65 mmHg. Balanced crystalloids and norepinephrine, which is the first-line vasopressor at our institution to treat intraoperative hypotension, were given at the discretion of the attending anesthesiologist.

For continuous finger-cuff arterial pressure monitoring, we used the ClearSight system (Edwards Lifesciences, USA). The ClearSight system has been validated against invasive arterial pressure measurements with an arterial catheter in patients having noncardiac surgery.^13–15  The appropriate finger-cuff (small, medium, or large) was positioned on the middle phalanx of the third or fourth finger. The heart reference sensor was attached to the thorax at the level of the right atrium, and the ClearSight system was zeroed before measurements began. Continuous beat-to-beat finger-cuff arterial pressure measurements were averaged automatically over nonoverlapping 20-s intervals and extracted from the ClearSight system.

For intermittent oscillometric arterial pressure monitoring, we used standard or large upper-arm cuffs (quimedic; Germany) connected to the patient monitor. Upper-arm cuffs were positioned on the arm contralateral to the finger-cuff. Oscillometric arterial pressures were initially measured at 2.5-min intervals, but clinicians were free to change the intervals, to cancel single measurements, or to perform additional measurements.

Hypotension exposure within 15 min after starting induction of anesthesia and during surgery was quantified using continuous finger-cuff arterial pressure data in all patients, whether assigned to continuous finger-cuff or assigned to intermittent oscillometric monitoring. We excluded finger-cuff arterial pressure values as artifacts using the following sequential rules: (1) arterial pressure values documented as artifacts by trial personnel; (2) systolic arterial pressures greater than 280 mmHg or less than 30 mmHg; (3) systolic arterial pressures less than diastolic arterial pressures plus 5 mmHg; or (4) diastolic arterial pressures greater than 150 mmHg or less than 10 mmHg. Excluded arterial pressure values were replaced by the mean of the closest arterial pressure values.

The trial had two hierarchical primary endpoints. The first primary endpoint was hypotension within 15 min after starting induction of anesthesia quantified as the area under a MAP of 65 mmHg (mmHg × min). The second primary endpoint was the time-weighted average MAP less than 65 mmHg (mmHg) during surgery. We chose a MAP of 65 mmHg to define hypotension because the intraoperative population harm threshold for organ injury is a MAP of 60 to 70 mmHg.^1,5 

Secondary endpoints within 15 min after starting induction of anesthesia included areas under MAP values of 60, 50, and 40 mmHg; durations of MAP values less than 65, 60, 50, and 40 mmHg; absolute and relative numbers of patients with any MAP less than 65, 60, 50, and 40 mmHg; absolute and relative numbers of patients with at least 1 consecutive min of a MAP less than 65, 60, 50, and 40 mmHg; areas above MAP values of 100, 110, 120, and 140 mmHg; norepinephrine dose; and norepinephrine use.

Secondary endpoints during surgery included time-weighted averages of MAP values less than 60, 50, and 40 mmHg; absolute and relative numbers of patients with any MAP less than 65, 60, 50, and 40 mmHg; absolute and relative numbers of patients with at least 1 consecutive min of a MAP less than 65, 60, 50, and 40 mmHg; time-weighted averages of MAP values above 100, 110, 120, and 140 mmHg; norepinephrine dose; and norepinephrine use.

Categorical data are presented as absolute numbers and percentages. Continuous data are presented as means ± standard deviations, medians with interquartile range, and ranges. Baseline imbalances between the two trial groups are described using absolute standardized differences, which are defined as the absolute value of the difference among means, mean rankings, or proportions divided by the pooled SD. We considered factors with absolute standardized differences exceeding 0.2 as imbalanced.

We conducted the primary endpoint analyses according to the intention-to-treat principle in full analysis sets (i.e., within 15 min after starting induction of anesthesia and during surgery). Patients in whom finger-cuff arterial pressure measurements were not available could not be analyzed. The first primary endpoint was analyzed using a two-sample Wilcoxon rank-sum test with a corresponding 95% CI at a 5% significance level. After confirming that the P value of the first primary endpoint was less than 0.05, we evaluated the secondary primary endpoint also at a 5% significance level using a two-sample Wilcoxon rank-sum test with a corresponding 95% CI. Continuous secondary endpoints were analyzed using a two-sample two-sided Wilcoxon rank-sum test with continuity correction; categorical secondary endpoints were analyzed using a Pearson’s chi-square test with Yates’s continuity correction.

We assumed that the area under a MAP of 65 mmHg within 15 min after starting induction of anesthesia would have a SD of 12 mmHg × min. The difference between a MAP of 65 mmHg (the routine lower MAP intervention threshold at our institution) and a MAP of less than 60 mmHg (at which the risk for postoperative complications significantly increases) is equal to a difference in an area of 5 mmHg × min (considering 1-min intervals). We thus considered a difference in the area under a MAP of 65 mmHg between the groups of 5.5 mmHg × min as minimal clinically important effect. To detect this clinically important difference of 5.5 mmHg × min or larger between the two groups with a power of 90% and a 5% significance level using a two-sided Wilcoxon rank-sum test, a total of 202 patients (n = 101 patients per group) are required. We expected a drop-out rate of 20% and thus planned to enroll a total of 242 patients, with 121 patients assigned to each group.

We enrolled and randomized 242 patients. Patients in whom finger-cuff arterial pressure measurements were not available were excluded from the full analysis sets because the primary endpoint could not be assessed (fig. 1). Finally, 222 patients were available for analysis of hypotension within 15 min after starting induction of anesthesia, and 227 were available for analysis of hypotension during surgery. Patient characteristics were well balanced between the two groups (table 1; supplementary table 1, https://links.lww.com/ALN/D161). We excluded 30 artifactual measurements among 10,585 total arterial pressure measurements (0.28%) within 15 min after starting induction of anesthesia and 254 among 52,099 (0.49%) during surgery.

Within 15 min after starting induction of anesthesia, the median (interquartile range) area under a MAP of 65 mmHg was 7 (0, 24) mmHg × min in patients assigned to continuous finger-cuff monitoring and 19 (0.3, 60) mmHg × min in patients assigned to intermittent oscillometric monitoring (P = 0.004; estimated location shift: −6 [95% CI: −15 to −0.3] mmHg × min; table 2; fig. 2).

During surgery, which lasted an average of 75 ± 39 min, the median (interquartile range) time-weighted average MAP less than 65 mmHg was 0.04 (0, 0.27) mmHg in patients assigned to continuous finger-cuff monitoring and 0.40 (0.03, 1.74) mmHg in patients assigned to intermittent oscillometric monitoring (P < 0.001; estimated location shift: −0.17 [95% CI: −0.41 to −0.05] mmHg; table 3; fig. 3).

Patients assigned to continuous finger-cuff monitoring also had smaller areas under MAP values of 60, 50, and 40 mmHg within 15 min after starting induction of anesthesia and lower time-weighted averages of MAP values less than 60, 50, and 40 mmHg during surgery (tables 2 and 3; figs. 2 and 3) than patients assigned to intermittent oscillometric monitoring. Furthermore, the number of patients with at least one 1-min episode of a MAP less than 65, 60, 50, and 40 mmHg was lower, and durations of MAP values less than 60, 50, and 40 mmHg were shorter in patients assigned to continuous finger-cuff—rather than intermittent oscillometric—monitoring (tables 2 and 3; supplementary table 2, https://links.lww.com/ALN/D162, and 3, https://links.lww.com/ALN/D163).

The median (interquartile range) norepinephrine dose within 15 min after starting induction of anesthesia was 0.02 (0.01 to 0.04) µg kg⁻¹ min⁻¹ in patients assigned to continuous finger-cuff monitoring compared to 0.01 (0 to 0.02) µg kg⁻¹ min⁻¹ in patients assigned to intermittent oscillometric monitoring (table 2). The median (interquartile range) norepinephrine dose during surgery was 0.05 (0.01 to 0.08) µg kg⁻¹ min⁻¹ in patients assigned to continuous finger-cuff monitoring compared to 0.02 (0 to 0.05) µg kg⁻¹ min⁻¹ in patients assigned to intermittent oscillometric monitoring (table 3).

Continuous finger-cuff arterial pressure monitoring helped clinicians substantially reduce hypotension within 15 min after starting induction of anesthesia and during noncardiac surgery compared to intermittent oscillometric monitoring. Within 15 min after starting induction of anesthesia, the area under a MAP of 65 mmHg was more than 60% smaller, and during surgery, the time-weighted average MAP less than 65 mmHg was about 90% lower in patients assigned to continuous finger-cuff monitoring than in those assigned to intermittent oscillometric monitoring.

Both severity and duration of hypotension are progressively associated with postoperative complications and mortality.¹⁶  It thus seems reasonable to intervene early when hypotension occurs. We therefore used area and time-weighted average below thresholds to quantify hypotension because both measures are based on severity and duration of hypotension. Time-weighted average is simply area divided by time and is thus normalized to surgical duration. Consequently, we report time-weighted averages during surgery because the duration of surgery varies but areas during induction of anesthesia because the observation period was 15 min in all patients.

Because organ injury presumably accrues at hypotensive extremes, we also considered cumulative durations and fractions of patients with at least one 1-min episode below various MAP thresholds. For instance, none of our patients assigned to continuous finger-cuff monitoring experienced a MAP less than 40 mmHg for more than 1 consecutive min within 15 min after starting induction of anesthesia, and only a single patient experienced a MAP less than 40 mmHg for more than 1 consecutive min during surgery. In contrast, four patients assigned to intermittent oscillometric monitoring had a MAP less than 40 mmHg for more than 1 consecutive min within 15 min after starting induction of anesthesia, and three patients did so during surgery. Continuous finger-cuff arterial pressure monitoring thus helped clinicians avoid severe hypotension that is presumably most dangerous.

There are two previous trials on continuous finger-cuff arterial pressure monitoring during surgery. In a trial with 160 patients having orthopedic surgery, using continuous finger-cuff versus using intermittent oscillometric arterial pressure monitoring more than halved the number of hypotensive events defined as a MAP less than 60 mmHg (19 vs. 51 events).¹⁷  In another trial with 316 patients having moderate to high-risk noncardiac surgery, using continuous finger-cuff arterial pressure monitoring nearly halved the amount of intraoperative hypotension defined as the time-weighted average MAP less than 65 mmHg (0.05 vs. 0.11 mmHg).¹⁸  Our results are thus consistent with both previous trials and add to the evidence that continuous finger-cuff arterial pressure monitoring helps clinicians reduce hypotension not only during noncardiac surgery but also within 15 min after starting induction of anesthesia.^17,18 

Our results are also consistent with a trial in which 242 noncardiac surgery patients were randomized to continuous intra-arterial or intermittent oscillometric arterial pressure monitoring during the first 15 min of induction of anesthesia.⁶  In this trial, there was only a third the area under a MAP of 65 mmHg with continuous intra-arterial arterial pressure monitoring than with intermittent oscillometric arterial pressure monitoring (15 vs. 46 mmHg × min).⁶  We do not propose that continuous finger-cuff monitoring should replace intra-arterial arterial pressure monitoring. However, clinicians might reasonably consider continuous finger-cuff monitoring in patients who would otherwise be monitored with intermittent oscillometric monitoring, and doing so will likely help reduce hypotension during induction of anesthesia and during surgery.

There are at least three reasons that continuous finger-cuff arterial pressure monitoring might help clinicians reduce hypotension. The first is that intermittent measurements naturally miss some episodes of hypotension and recognize others belatedly. A second reason is that oscillometry overestimates low arterial pressures and may thus miss hypotension.^19–21  Finally, the measurement performance of finger-cuff methods may be better than that of oscillometry. For example, we recently reported that the agreement between finger-cuff and intra-arterial arterial pressures was better than that between oscillometric and intra-arterial arterial pressures in obese patients having surgery.¹⁵ 

Continuous monitoring per se will not prevent hypotension unless it triggers effective interventions. Consistent with this theory, patients assigned to continuous finger-cuff monitoring were given more than twice as much norepinephrine both within 15 min after starting induction of anesthesia and during surgery than patients assigned to intermittent oscillometric monitoring, presumably explaining why there was substantially less hypotension.

This trial has limitations. We quantified hypotension based on continuous finger-cuff arterial pressure monitoring in all patients because previous studies suggest that the agreement between finger-cuff and intra-arterial arterial pressures is noninferior or even better than the agreement between oscillometric and intra-arterial arterial pressures.^15,22  Doing so eliminated the potential influence of measurement differences between finger-cuff and oscillometric arterial pressure measurements on our results. However, finger-cuff monitoring systems vary in accuracy,⁹  and results will differ at least somewhat with other finger-cuff systems.

We did not use a specific hemodynamic treatment protocol and thus evaluated only the effect of arterial pressure monitoring on hypotension. Our trial was not designed to investigate whether the reduction in hypotension by using continuous finger-cuff arterial pressure monitoring also reduces postoperative complications, but the results of our trial provide a solid basis for designing future large-scale trials with hard outcomes.

This was a single-center trial. The amount of hypotension within 15 min after starting induction of anesthesia and during surgery will presumably differ in other centers and in other patients. Also important is that the effect of continuous arterial pressure monitoring on arterial pressure depends critically on clinician responses. In addition, the amount of hypotension reduction may have been influenced by the fact that finger-cuff monitoring is not routinely used and thus may have increased clinicians’ attention for hypotension. Nonetheless, hypotension was both common and profound.

We separately considered hypotension within 15 min after starting induction of anesthesia and during surgery because anesthetic induction is performed in a central induction area in our institution. After anesthetic induction, patients are moved into the operating room, positioned, and prepared for surgery. Movement often causes measurement artifacts. We therefore separately considered the initial 15 min after starting induction of anesthesia and the intraoperative period.

In summary, continuous finger-cuff arterial pressure monitoring helps clinicians reduce hypotension within 15 min after starting induction of anesthesia and during noncardiac surgery compared to intermittent oscillometric arterial pressure monitoring. A much larger trial would be needed to determine whether the observed hypotension reduction results in a reduction of perfusion-related complications, but in the meantime, clinicians can use continuous finger-cuff instead of intermittent oscillometric arterial pressure monitoring to reduce hypotension during induction of anesthesia and during surgery.

Supported by Edwards Lifesciences (Irvine, California), which provided the EV1000 monitoring system and finger-cuff sensors for the trial. Edwards Lifesciences was not involved in the development of the trial design, data acquisition or analysis, writing of the manuscript, or the decision to submit the manuscript for publication.

Dr. Kouz is a consultant for and has received honoraria for giving lectures from Edwards Lifesciences (Irvine, California) and is a consultant for Vygon (Aachen, Germany). Dr. Petzoldt has received institutional restricted research grants from Verathon Inc. (Bothell, Washington). Dr. Sessler is a consultant for Edwards Lifesciences and has received research funding from the company; he also has equity interests in Perceptive Medical (Newport Beach, California). Dr. Flick has received honoraria for giving lectures from Edwards Lifesciences and has received institutional restricted research grants and honoraria for giving lectures and consulting from CNSystems Medizintechnik GmbH (Graz, Austria). Dr. Saugel is a consultant for and has received institutional restricted research grants and honoraria for giving lectures from Edwards Lifesciences; he is also a consultant for and has received institutional restricted research grants and honoraria for giving lectures from Pulsion Medical Systems SE (Feldkirchen, Germany). Dr. Saugel has received honoraria for giving lectures from Getinge AB (Gothenburg, Sweden). Dr. Saugel has received institutional restricted research grants and honoraria for giving lectures from CNSystems Medizintechnik GmbH. BS is a consultant for and has received honoraria for giving lectures from Philips Medizin Systeme Böblingen GmbH (Böblingen, Germany). Dr. Saugel is a consultant for and has received institutional restricted research grants and honoraria for giving lectures from GE Healthcare (Chicago, Illinois); is a consultant for and has received honoraria for giving lectures from Vygon; is a consultant for and has received honoraria for giving lectures from Baxter (Deerfield, Illinois); is a consultant for and has received institutional restricted research grants from Retia Medical LLC. (Valhalla, New York); has received institutional restricted research grants from Osypka Medical (Berlin, Germany); and was a consultant for and has received institutional restricted research grants from Tensys Medical Inc. (San Diego, California). The other authors declare no competing interests.

Full protocol available at: bernd.saugel@gmx.de.

Supplementary Table 1: Summary of additional clinical characteristics (n = 231), https://links.lww.com/ALN/D161

Supplementary Table 2: Summary of additional secondary outcomes within 15 min after starting induction of anesthesia (n = 222), https://links.lww.com/ALN/D162

Supplementary Table 3: Summary of additional secondary outcomes during surgery (n = 227), https://links.lww.com/ALN/D163