Introduction
The perianesthetic mortality rate is higher in animals,1 compared with humans,2 and the reason for this observation is multifactorial. One possible explanation is that the quality and quantity of patient monitoring equipment are lower in an average veterinary practice versus in human hospitals.3 For example, of 487 dogs anesthetized in veterinary practices in the United Kingdom, a PO was used intraoperatively for only 51%, an ECG monitor for only 11%, a capnographic monitor for only 10%, and a blood pressure monitor for only 10%.4 Surveys of veterinarians in the United Kingdom3 and New Zealand5 reveal that a substantial proportion of anesthetized animals are not monitored with any electronic devices. For anesthetized cats, Spo2 and PR monitoring is associated with decreased odds of anesthetic death,6 and the absence of entries of intra-anesthetic Spo2 measurements in their medical records is associated with increased odds of death.1 With 74% of perianesthetic deaths associated with cardiovascular or respiratory causes in a study7 of small animals, the importance of monitoring the function of these systems during anesthesia is further emphasized. Additionally, a dedicated anesthetist may only be intermittently available for patient monitoring in many veterinary practices. In such situations, continuous audible monitoring of the pulse or heart rate is advised.8,a Although multiple options (eg, Doppler ultrasonic flow sensor or ECG) to achieve this goal are available, pulse oximetry with placement of a sensor on the patient's tongue is probably the simplest and quickest option. Pulse oximetry is not recommended to be the sole monitoring modality, yet it may be the most commonly used one.4
Pulse oximetry calculates the O2 saturation of arterial blood and PR via continuous measurement of the relative absorption of red and infrared light beams directed through a perfused tissue bed during multiple cardiac cycles. Recently, portable, compact, battery-operated, cableless FPOs have become popular for aviators and for recreational and medical use in people.9 Although these FPOs use technology similar to that of the medical POs, many are not FDA approved for medical use9 and none are validated for veterinary use. Most FPOs cost < 10% of the price of medical-grade POs9 and therefore may reduce the financial barrier to increasing the quantity of affordable patient monitoring equipment in veterinary practices.3 Also, portability of FPOs may allow more veterinarians to use a PO for monitoring anesthetized patients not only in a clinical setting but also in out-of-hospital emergency and field anesthesia settings. They may also be useful during anesthetic recovery or in the intensive care unit, where animals may damage large expensive devices. Despite potential advantages of FPOs, their use with veterinary patients has not yet been reported. Therefore, the objective of the study reported here was to evaluate the repeatability and accuracy for Spo2 and PR measurements obtained with various FPOs in anesthetized dogs.
Materials and Methods
Animals
This study was approved by the University of Illinois Institutional Animal Care and Use Committee (protocol No. 18246), and informed owner consent was obtained prior to a dog's inclusion in the study. The criteria for inclusion were the following: client-owned dogs presenting to the veterinary teaching hospital for various procedures requiring general anesthesia for > 30 minutes, physical status classification of 1 to 2 according to the American Society of Anesthesiologists, body weight ≥ 10 kg, and an intact nonpigmented tongue. Dogs were excluded if they had any condition that may have compromised the respiratory system (eg, obesity or pregnancy); had general anesthesia for CT, MRI, or radiation therapy; had surgery of the head or neck; had any situation that limited accessibility to the tongue; or required intensive medical attention. Furthermore, dogs were excluded if any arrhythmia, including sinus arrhythmia, was present; mean arterial blood pressure was ≤ 60 mm Hg or ≥ 120 mm Hg; core body temperature was ≤ 36°C or ≥ 39°C; end-tidal partial pressure of CO2 was ≤ 35 mm Hg or ≥ 60 mm Hg; dogs required repositioning or IV administered medications during data collection; or an α2-adrenoceptor agonist had been administered 15 minutes prior to or during the time of data collection. On the basis of the investigators’ observations, a period of 15 minutes was commonly sufficient for stabilization of α2-adrenoceptor agonist–induced cardiovascular changes. With the assumptions that a 2% difference in bias between FPOs was clinically significant and a 2% SD was expected for each FPO on the basis of preliminary data (not shown), power was 95%, and α was 0.05, these data inputted into a software programb resulted in the need for a minimum sample size of 27 dogs to compare 2 independent means.10 Considering potential attrition, the goal was to recruit 30 dogs.
Experimental protocol
The attending anesthesiologist determined the anesthetic protocol for each dog. Common protocols included premedication with an opioid with or without dexmedetomidine, induction with propofol with or without lidocaine or ketamine, and maintenance with isoflurane. After orotracheal intubation, the endotracheal tube cuff was inflated and each dog breathed 100% O2 for a minimum of 10 minutes before beginning the measurement series.
To determine the PR at the beginning of the measurement series for each dog, a peripheral artery (eg, one of the digital, dorsal pedal, labial, or lingual arteries) was palpated and the PR counted over 30 seconds. All dogs had a regular sinus rhythm, and the heart rate as determined with the simultaneously displayed ECG on the anesthesia monitorc matched with the PR at the initial assessment, such that the heart rate displayed by the monitor thereafter was used as the reference PR. A > 5% change in the reference PR was confirmed for accuracy by palpation of the previously used peripheral artery.
Nine FPOs were purchased from an online vendord in November 2018 by use of search criteria as follows: ≥ 1,000 product reviews and average product ratings of ≥ 4 out of 5 stars. This vendor was chosen because it is one of the largest online marketplaces in the United States, and the search criteria ensured that the chosen FPOs had the highest sales volume and customer satisfaction ratings. Two models were excluded after purchase because one had no identifiable manufacturer reference and the other was a duplicate model; therefore, 7 FPO models (FPO 1 through 7)e–k remained in the study. The reference for Spo2 was determined with a medical-grade handheld PO.l
Eight POs (7 FPOs and the reference PO) were evaluated in random orderm for each dog. Investigators were blinded to the test order prior to determining dog eligibility. Each PO was tested twice in rapid sequence, with removal and replacement of the device between each recording. The sensor of the first of 8 POs was placed on the tip of the tongue, parallel to its midline. The PR and Spo2 measurements were obtained starting 20 seconds after placement of the PO sensor, and recording of Spo2 measurements was immediately followed by recording the reference PR from the anesthesia monitor. If no values were displayed at 20 seconds, investigators waited for up to 60 seconds for values to display. If no values appeared by this time, the measurement was aborted. Preliminary investigations revealed that 20 seconds was sufficient for most POs to provide stable readings. An end time of 60 seconds was arbitrarily chosen because a longer response time was considered to be impractical for clinical use. After obtaining the first measurements from the first PO, the device was briefly removed then immediately repositioned to the same location on the tongue as before and all measurements were repeated. Thereafter, the sensor of the first PO was removed, and the sensor for the next PO was similarly placed and the same measurement sequence was repeated. Measurements on each dog were performed by 1 of 6 investigators (a board-certified anesthesiologist [TDA and SCJK], anesthesiology resident [DESH, AMS, and FDCM], or anesthesiology intern [SCD)]); all investigators were trained to perform the study protocol with a uniform technique. Investigators had no influence on clinical decision-making and management of any dog. All measurements on a single dog were completed in approximately 20 to 30 minutes.
Statistical analysis
Normal distribution of the data could not be assumed with the Shapiro-Wilk test. Therefore, nonparametric descriptive statistics were reported.
Median differences between duplicate Spo2 measurements were calculated for each PO, and the interval between the 5th and 95th percentiles were used to define CR. Acceptable values for CR were ≤ 6% Spo2 on the basis that a maximum ± 3% difference was acceptable. The number of times an FPO erroneously measured Spo2 or PR was tabulated. An erroneous measurement was defined as a duplicate measurement with an unacceptable CR or a PO that did not provide a reading. Coefficient of repeatability for PR was not determined for any PO because the reference PR value was not always stable during duplicate measurements, which would have been a prerequisite for assessing repeatability.
The first of each duplicate FPO measurement was compared with that of the reference PO on the basis of nonparametric statistics by Bland-Altman analysis (ie, assessing agreement with the reference method).11 Absolute differences in Spo2 and relative differences in PR (ie, percentage difference from the reference PR) between FPO and reference PO measurements were calculated. Median of the differences was regarded as bias and the 5th and 95th percentiles as LOAs. The a priori LOAs were arbitrarily defined as a ± 3% absolute difference in Spo2 or ± 10% relative difference in PR, and the agreement was acceptable when the LOAs were equal to or within these limits; therefore, in such cases, measurements obtained with an FPO were considered accurate. Commercially available softwaren was used for statistical analysis.
Results
Thirty dogs were enrolled in the study. One dog was excluded after it was anesthetized because its tongue was pigmented. Thus, 29 dogs completed the study. Mean (SD) age was 3.8 (2.4) years and body weight was 33.0 (13.0) kg. Eighteen dogs were female (sexually intact, n = 3; spayed, 15) and 11 dogs were male (sexually intact, 2; neutered, 9). Breeds of dogs included Labrador Retriever (n = 7), American Pit Bull Terrier (2), American Staffordshire Terrier (2), Bernese Mountain Dog (2), Golden Retriever (2), and Akita, Belgian Malinois, Cavalier King Charles Spaniel, Cocker Spaniel, Goldendoodle, Neapolitan Mastiff, and Standard Poodle (1 each); there were also 7 mixed-breed dogs. Dogs were anesthetized for procedures as follows: tibial plateau leveling osteotomy (n = 19), bilateral elbow arthroscopy, femoral head and neck ostectomy, forelimb amputation, gastropexy, tarsal arthrodesis, laparoscopic-assisted ovariohysterectomy plus gastropexy, medial patellar luxation stabilization, ovariohysterectomy, ovariectomy, and tibial fracture repair (1 each).
With an expected 232 paired measurements, 231 were obtained, with only 1 measurement not possible because the battery failed for one of the FPOs. Five pairs of measurements were excluded from repeatability analysis and 3 from Bland-Altman analysis because the FPOs did not provide readings. Hence, 226 paired measurements were included in the repeatability analysis, and 228 were included in the comparison analysis. The median (range) of the reference Spo2 was 100% (99% to 100%) and PR was 71 beats/min (38 to 100 beats/min).
Results of repeatability (ie, precision) and Bland-Altman analyses are summarized (Table 1). Median difference between repeated Spo2 measurements with each FPO or the reference PO was 0, and CRs were acceptable for all devices at ≤ 6%, except for FPOs 5i and 7.k All FPOs underestimated Spo2 with a median bias ranging between −1% and −2%, and LOAs were wider than ± 3%. Bias for PR was 0 for all POs, and LOAs were within ± 10% for FPOs 2f and 4h and the PO. The proportions of erroneous measurements for Spo2 were ≤ 3 of 29 for repeatability analysis with FPOs 1,e 2,f 3,g 4,h and 6j and for the PO, which happened to coincide with acceptable CR measurements but not for Bland-Altman analysis, in which the error proportions were > 7 of 29 for all FPOs (Table 2). However, the proportions of erroneous measurements for PR (Bland-Altman analysis) were ≤ 3 of 29 for FPOs 1,e 2,f 4,g 5,i and 6j and for the PO, which were similar, but not identical, to results obtained with evaluation of the LOAs.
Results of repeatability and Bland-Altman analyses for Spo2 and PR measurements by 7 FPOs,e–k 1 PO,l and the anesthesia monitorc (PR only) during general anesthesia of 29 dogs that underwent surgery for various reasons (eg, orthopedic problems and ovariohysterectomy) and were breathing 100% O2.
| PO type | Repeatability | Bland-Altman | |||
|---|---|---|---|---|---|
| No. | Spo2 | No. | Spo2 | PR | |
| FPO 1†,e | 27 | 0 (5.1)* | 28 | −1 (−6, 0) | 0 (−7, 14) |
| FPO 2f | 29 | 0 (4.2)* | 29 | −2 (−5, 1) | 0 (−8, 7)* |
| FPO 3g | 28 | 0 (4.5)* | 28 | −1 (−5, 0) | 0 (−6, 23) |
| FPO 4†,h | 28 | 0 (3.2)* | 28 | −1 (−4, 0) | 0 (−7, 8)* |
| FPO 5†,i | 28 | 0 (6.7) | 29 | −2 (−7, 0) | 0 (−11, 6) |
| FPO 6†,j | 29 | 0 (5.0)* | 29 | −2 (−5, 0) | 0 (−3, 12) |
| FPO 7k | 28 | 0 (6.1) | 28 | −2 (−6, 0) | 0 (−4, 27) |
| PO†,j | 29 | 0 (2.0)* | 29 | — | 0 (−4, 5)* |
Median (CR) absolute differences for Spo2 (%) between the 2 consecutive readings for the same FPO or PO indicate degree of repeatability (ie, precision). Median (5th and 95th percentiles) differences for Spo2 between each FPO and the reference PO indicate bias (LOA) as determined by Bland-Altman analysis. Similarly, median (5th and 95th percentiles) differences for PR between each PO and the reference (anesthesia monitor and occasionally by palpation of a peripheral artery), expressed as percentage of the reference PR, indicate bias (LOA).
Indicates acceptable repeatability (Spo2 CR ≤ 6%) or agreement (Spo2 LOA < ± 3% or PR LOA < ± 10%).
FDA-approved.
= Not applicable. No. = Number of measurements attained within 60 seconds of FPO or PO placement.
Proportion (%) of the number of erroneous measurements (Spo2 > ± 3%; PR > ± 10% or PO did not display a reading) for repeatability (ie, precision) and Bland-Altman analyses for the dogs of Table 1.
| PO type | Repeatability | Bland-Altman | ||
|---|---|---|---|---|
| Spo2 | Spo2 | PR | ||
| FPO 1†,e | 3/29 (10) | 8/29 (28) | 3/29 (10) | |
| FPO 2f | 1/29 (3) | 8/29 (28) | 2/29 (7) | |
| FPO 3g | 3/29 (10) | 7/29 (24) | 4/29 (14) | |
| FPO 4†,h | 2/29 (7) | 7/29 (24) | 2/29 (7) | |
| FPO 5†,i | 7/29 (24) | 13/29 (45) | 1/29 (3) | |
| FPO 6†,j | 2/29 (7) | 11/29 (38) | 1/29 (3) | |
| FPO 7k | 4/28 (14) | 8/28 (29) | 3/28 (11) | |
| PO†,l | 0/29 (0) | — | 1/29 (3) | |
See Table 1 for key.
Discussion
The present study revealed the repeatability (ie, precision) and accuracy for 7 FPOs used with anesthetized dogs breathing 100% O2. Repeatability of Spo2 measurements was acceptable for 5 of 7 FPOs, but all underestimated Spo2 versus the reference PO. Surprisingly, the PR results for 2 FPOs (FPOs 2f and 4h) agreed with the reference method (anesthesia monitor or palpation of peripheral artery), with LOA defined as a difference of ≤ ± 10% between measurements; therefore, these 2 FPOs were considered accurate for PR measurement. However, if a difference of ≤ ± 20% in PR measurements was instead considered acceptable, then 5 FPOs were suitable for PR measurements. These data suggested that some of the FPOs may be useful for accurately monitoring the PR of anesthetized dogs but not the Spo2.
A study9 of conscious people indicates that 4 of 6 FPOs had high mean bias for Spo2 that ranged from 70% to 100% (up to −6.3%) and wide LOAs (vs 2 hemoximeters). One of the FPOs used in that study9 was also included in the present study (FPO 1e), but it was accurate with an overall bias (LOA) of 0.6% (range, −2.9% to 4.2%). Another study12 reveals that the same FPO (FPO 1e in the present study) had a bias (LOA) of −0.2% (−2.3% to 2.2%) for Spo2 (vs standard PO) in people receiving perioperative care. Yet, direct comparison of those findings with results of the present study is difficult because the statistical analyses (ie, parametric or nonparametric) differ.
However, the LOA for each FPO evaluated in the present study was wider than those reported in the aforementioned studies,9,12 with the differences most likely because these FPOs were designed for use with human fingers, not with dog tongues. An FPO's sensor is designed to align with the tip of a person's fingernail, and a dog's tongue may not always fully align. Visual confirmation of a tongue's position was not possible. Malalignment may lead to a low signal-to-noise ratio, loss of signal trace, and reduced Spo2 measurements resulting from various sources of errors, such as optical shunting (ie, a part of the PO sensor's light beam passes through without interacting with tissues), low tissue blood volume (eg, vasoconstriction or hypovolemia), or arteriovenous anastomoses (found in the tip of a dog's tongue).13,14 One study15 of dogs reveals that Spo2 measurements are affected by the location of the PO sensor on the tongue, with higher Spo2 measurements obtained when the sensor is positioned at the center of the tongue versus laterally or at the base. Yet, a recent study16 of dogs reveals no difference in Spo2 measurements with sensor positioning at the lateral aspects of or partially off the tongue, causing 50% optical shunting. Although the tip of the tongue has the most arteriovenous anastomoses in dogs,17 neither study15,16 includes a comparison of Spo2 measurements obtained with the sensor placed at the tip versus the center of the tongue. Because the tip of the tongue was the chosen site for sensor placement of the FPOs evaluated in the present study, that may have negatively affected the accuracy of Spo2 measurements. Nevertheless, the tip of the tongue remains the preferred site for FPO placement in dogs because sensor alignment may be difficult from a lateral approach (depending on an individual dog's tongue anatomy and the construction of an FPO probe) and positioning of the sensor over the central area of the tongue is not possible. During preliminary investigations, use of FPOs with cats was attempted, but the FPOs did not fit on their tongues and readings were rare with sensor placement on a foot.
Most POs are sensitive to errors associated with skin pigmentation,18,19 ambient light,20,21 extreme temperatures,22,23 sensor malalignment,24,25 motion,21,26 low perfusion states,27 and tissue compression by the sensor.15 Similar limitations were also expected for FPOs, especially when used with animals. In the present study, dogs that had pigmented tongues or abnormal body temperature or blood pressure were excluded. A study15 of dogs and cats reveals that differences in contact pressure on the tongue affects the accuracy of Spo2 measurements. To minimize potential inaccuracy resulting from tissue compression, test order of FPOs was randomized in the present study.
The reference PO used in the present study was equipped with signal processing technology that maintains accuracy and precision at different O2 saturation levels in human patients in the intensive care unit.28 Also, the reference PO performs favorably with cynomolgus monkeys29 and rats.30 A co-oximeter, the gold standard for Spo2 measurements, would have been the optimal reference method, but one was not available for use in the present study. However, comparing the FPOs with the reference PO was still considered appropriate for determining whether the FPOs were interchangeable with the reference PO under the study conditions.
Recommendations for evaluating POs in people by the ISO31 are accepted by the FDA.32 The guidelines recommend comparing Spo2 measurements to co-oximeter measurements at different evenly distributed O2 saturation plateaus ranging from 70% to 100% and examining at least 200 data points/PO. The ISO suggests that an acceptable root mean square difference between methods shall be ≤ 4%.31 The FDA further specified that for clip sensors, this value should be ≤ 3%.32 In the present study, following these guidelines was not possible because the experimental design differed from the ISO guidelines and normal distribution of the data, a requirement to calculate root mean square differences, could not be assumed. Repeatability of Spo2 measurements for some of the FDA-approved FPOs (eg, FPO 5i) was below that recommended by the ISO. An important limitation of the present study was that a wide range of O2 saturation plateaus was not examined; therefore, results related to Spo2 should be interpreted cautiously. Evaluating FPOs at various O2 saturation plateaus would have been ideal but was not possible with patients. Nevertheless, anesthetized dogs are likely to be supplemented with O2, and learning whether FPOs provide accurate measurements of Spo2 and PR under these common circumstances is important. Neither the ISO nor the FDA guidelines include recommendations for assessing the accuracy of PR measurements. Therefore, the PR limit (≤ ± 10%) used in the present study was adopted on the basis of a previous study.33
Another limitation of the present study was the use of healthy dogs. Pathological conditions, such as low perfusion states, may influence the accuracy of POs34; therefore, the accuracy reported here may be different with ill dogs. Additionally, the FPOs were used sporadically rather than continuously. Over time, accuracy tended to improve for many FPOs, and it may have been different than that reported here if more time (> 60 seconds) had elapsed before obtaining measurements. A further complicating factor in the present study was that the sensors were positioned on the tongue at the same location one after another and the order may have influenced the perfusion of the tongue and, hence, the accuracy of the FPOs. As for mitigating the possibility of systematic errors resulting from tissue manipulation, test order of FPOs was randomized in the present study. Several other factors may have influenced measurement accuracy, such as differences in drug protocols, time between induction of anesthesia and data acquisition, surgical procedures, and patient position, age, sex, and breed. However, these factors were randomly distributed and likely affected all FPO measurements similarly.
The use of a PO is associated with a reduced anesthetic mortality rate in cats,1,6 but whether monitoring of Spo2 or PR contributed to this reduced rate is unknown. Yet, because most cats in those studies1,6 received 100% O2, intraoperative O2 desaturation was likely rare and more likely caused by a low PR (severe reduction over time) possibly displayed by a PO, subsequently leading to better patient care (ie, intervention as needed to mitigate low PR). Furthermore, whether to use a PO may have been influenced by the patient's condition; drugs administered for premedication and induction and maintenance of anesthesia; and whether an IV catheter was placed. Monitoring of PR may be more reliable with other devices (eg, Doppler ultrasonic flow sensor) rather than with a PO, but POs, if any device is used, are the most common monitoring devices in veterinary practice.3–5 On the basis of the present study, various low-cost FPOs that automatically monitor the PR were reasonably accurate with anesthetized dogs. Therefore, veterinarians may now have affordable and reasonably accurate options to monitor the PR of anesthetized dogs that may not have otherwise been monitored. Additionally, having a low-cost FPO may complement existing monitoring devices used in general practice.
Overall, the best performing FPOs for the anesthetized dogs breathing 100% O2 in the present study were FPOs 2f and 4,h such that they accurately measured PR (error, ≤ ±10%) and only mildly underestimated Spo2. The FPOs 1,e 5i, and 6j were less accurate (error, ≤ ± 20%) but may still be useful in practice. However, FPO 5i had outstandingly low precision and accuracy for Spo2 measurements, and FPOs 4h and 6j were no longer commercially available after completion of the present study. Therefore, FPOs 1e and 2f could be considered for further testing in veterinary research and clinical settings, including dogs in hypoxemic or low perfusion states.
Abbreviations
| CR | Coefficient of repeatability |
| FPO | Fingertip pulse oximeter |
| ISO | International Organization for Standardization |
| LOA | Limit of agreement |
| PO | Pulse oximeter |
| PR | Pulse rate |
| Spo2 | Oxygen saturation in arterial blood as measured by pulse oximetry |
Footnotes
ACVA Monitoring Guidelines Update, 2009, Lakewood, Colo. Available at: aaha.org/globalassets/02-guidelines/anesthe sia/8-9-16_acva_monitoring_guidelines.pdf. Accessed Feb 22, 2020.
G*Power, version 3.1.9.6, Dusseldorf University, Dusseldorf, Germany.
Passport 12 multi-parameter monitor, Mindray North America, Mahwah, NJ.
Amazon.com, Seattle, Wash.
CMS 50-DL, Contec Medical Systems Co Ltd, Qinhuangdao, Hebei, People's Republic of China.
Zacurate Pro series 500DL, Zacurate, Stafford, Tex.
SM-110, Santa Medical, Tustin, Calif.
SM-165, Santa Medical, Tustin, Calif.
Innovo Deluxe, Innovo Medical, Stafford, Tex.
Vive Precision, Vive Health, Naples, Fla.
BlackOx, Concord Health Supply, Skokie, Ill.
Nonin 8500, Nonin Medical Inc, Plymouth, Minn.
Research Randomizer, version 4.0, Middletown, Conn. Available at: randomizer.org. Accessed Apr 9, 2019.
Matlab, version R2019b, The MathWorks Inc, Natick, Mass.
References
- 1. ↑
Matthews NS, Mohn TJ, Yang M, et al. Factors associated with anesthetic-related death in dogs and cats in primary care veterinary hospitals. J Am Vet Med Assoc 2017;250:655–665.
- 2. ↑
Bainbridge D, Martin J, Arango M, et al. Perioperative and anaesthetic-related mortality in developed and developing countries: a systematic review and meta-analysis. Lancet 2012;380:1075–1081.
- 3. ↑
Richardson EM, McMillan M. Survey on conduct of anaesthetic monitoring in small animal practice in the UK. Vet Rec 2019;185:570.
- 4. ↑
Brodbelt DC, Pfeiffer DU, Young LE, et al. Results of the confidential enquiry into perioperative small animal fatalities regarding risk factors for anesthetic-related death in dogs. J Am Vet Med Assoc 2008;233:1096–1104.
- 5. ↑
Sano H, Barker K, Odom T, et al. A survey of dog and cat anaesthesia in a sample of veterinary practices in New Zealand. N Z Vet J 2018;66:85–92.
- 6. ↑
Brodbelt DC, Pfeiffer DU, Young LE, et al. Risk factors for anaesthetic-related death in cats: results from the confidential enquiry into perioperative small animal fatalities (CEPSAF). Br J Anaesth 2007;99:617–623.
- 7. ↑
Brodbelt DC, Blissitt KJ, Hammond RA, et al. The risk of death: the confidential enquiry into perioperative small animal fatalities. Vet Anaesth Analg 2008;35:365–373.
- 8. ↑
Grubb T, Sager J, Gaynor JS, et al. 2020 AAHA Anesthesia and Monitoring Guidelines for Dogs and Cats. J Am Anim Hosp Assoc 2020;56:59–82.
- 9. ↑
Lipnick MS, Feiner JR, Au P, et al. The accuracy of 6 inexpensive pulse oximeters not cleared by the Food and Drug Administration: the possible global public health implications. Anesth Analg 2016;123:338–345.
- 10. ↑
Faul F, Erdfelder E, Lang AG, et al. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175–191.
- 11. ↑
Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res 1999;8:135–160.
- 12. ↑
Smith RN, Hofmeyr R. Perioperative comparison of the agreement between a portable fingertip pulse oximeter v. a conventional bedside pulse oximeter in adult patients (COMFORT trial). S Afr Med J 2019;109:154–158.
- 14. ↑
Kelleher JF, Ruff RH. The penumbra effect: vasomotion-dependent pulse oximeter artifact due to probe malposition. Anesthesiology 1989;71:787–791.
- 15. ↑
Mair A, Martinez-Taboada F, Nitzan M. Effect of lingual gauze swab placement on pulse oximeter readings in anaesthetised dogs and cats. Vet Rec 2017;180:49.
- 16. ↑
Mair A, Ferreira J, Ricco C, et al. Appraisal of the ‘penumbra effect’ using lingual pulse oximetry in anaesthetized dogs and cats. Vet Anaesth Analg 2020;47:177–182.
- 17. ↑
Kishi Y, So S, Harada Y, et al. Three-dimensional SEM study of arteriovenous anastomoses in the dog's tongue using corrosive resin casts. Acta Anat (Basel) 1988;132:17–27.
- 18. ↑
Bickler PE, Feiner JR, Severinghaus JW. Effects of skin pigmentation on pulse oximeter accuracy at low saturation. Anesthesiology 2005;102:715–719.
- 19. ↑
Adler JN, Hughes LA, Vivilecchia R, et al. Effect of skin pigmentation on pulse oximetry accuracy in the emergency department. Acad Emerg Med 1998;5:965–970.
- 20. ↑
Fluck RR Jr, Schroeder C, Frani G, et al. Does ambient light affect the accuracy of pulse oximetry? Respir Care 2003;48:677–680.
- 21. ↑
Trivedi NS, Ghouri AF, Shah NK, et al. Effects of motion, ambient light, and hypoperfusion on pulse oximeter function. J Clin Anesth 1997;9:179–183.
- 22. ↑
Schramm WM, Bartunek A, Gilly H. Effect of local limb temperature on pulse oximetry and the plethysmographic pulse wave. Int J Clin Monit Comput 1997;14:17–22.
- 23. ↑
Langton JA, Lassey D, Hanning CD. Comparison of four pulse oximeters: effects of venous occlusion and cold-induced peripheral vasoconstriction. Br J Anaesth 1990;65:245–247.
- 24. ↑
Barker SJ, Hyatt J, Shah NK, et al. The effect of sensor malpositioning on pulse oximeter accuracy during hypoxemia. Anesthesiology 1993;79:248–254.
- 25. ↑
Southall DP, Samuels M. Inappropriate sensor application in pulse oximetry. Lancet 1992;340:481–482.
- 26. ↑
Goldman JM, Petterson MT, Kopotic RJ, et al. Masimo signal extraction pulse oximetry. J Clin Monit Comput 2000;16:475–483.
- 27. ↑
Hummler HD, Engelmann A, Pohlandt F, et al. Accuracy of pulse oximetry readings in an animal model of low perfusion caused by emerging pneumonia and sepsis. Intensive Care Med 2004;30:709–713.
- 28. ↑
Singh AK, Sahi MS, Mahawar B, et al. Comparative evaluation of accuracy of pulse oximeters and factors affecting their performance in a tertiary intensive care unit. J Clin Diagn Res 2017;11:OC05–OC08.
- 29. ↑
Young SS, Skeans SM, Lamca JE, et al. Agreement of Spo2, Sao2 and Sco2 in anesthetized cynomolgus monkeys (Macaca fascicularis). Vet Anaesth Analg 2002;29:150–155.
- 30. ↑
Bernard SL, An D, Glenny RW. Validation of the Nonin 8600V pulse oximeter for heart rate and oxygen saturation measurements in rats. Contemp Top Lab Anim Sci 2004;43:43–45.
- 31. ↑
ISO. ISO 80601-2-61:2017: medical electrical equipment—part 2–61: particular requirements for basic safety and essential performance of pulse oximeter equipment. Geneva: ISO, 2017.
- 32. ↑
Center for Devices and Radiological Health. Pulse oximeters - premarket notification submissions [510(k)s]: guidance for industry and Food and Drug Administration staff. Rockville, Md: FDA, 2013.
- 33. ↑
Nelson BW, Allen NB. Accuracy of consumer wearable heart rate measurement during an ecologically valid 24-hour period: intraindividual validation study. JMIR Mhealth Uhealth 2019;7:e10828.
- 34. ↑
Secker C, Spiers P. Accuracy of pulse oximetry in patients with low systemic vascular resistance. Anaesthesia 1997;52:127–130.
