The Efficacy of Perfusion Index as an Indicator for Intravascular Injection of Epinephrine-Containing Epidural Test Dose in Propofol-Anesthetized Adults : Anesthesia & Analgesia

Secondary Logo

Journal Logo

Patient Safety: Research Reports

The Efficacy of Perfusion Index as an Indicator for Intravascular Injection of Epinephrine-Containing Epidural Test Dose in Propofol-Anesthetized Adults

Mowafi, Hany A. MBBch, MSc, MD; Ismail, Salah A. MBBch, MSc, MD; Shafi, Mohammed A. MBBch, MSc, MD; Al-Ghamdi, AbdulMohsin A. MBBch, MD

Author Information
doi: 10.1213/ane.0b013e31818fc35b
  • Free
  • Chinese Language Editions

Regional analgesia is commonly combined with general anesthesia to reduce intraoperative general anesthetic requirements and provide postoperative pain relief.1,2 Administration of local anesthetics has a potential hazard of intravascular injection, inducing life-threatening central nervous system and cardiovascular toxicity.3,4 Therefore, for safer management of patients receiving local anesthetics during general anesthesia, an epidural test dose containing 15 μg epinephrine has been recommended.5–7

In addition to the conventional hemodynamic criteria, digital skin blood flow as measured by a laser Doppler flowmeter was found to be a reliable marker for intravascular test dose injection during inhaled anesthesia.8 However, the cost and availability of laser Doppler flowmeters may limit the routine use of this diagnostic method. Plethysmographic pulse wave monitoring, on the other hand, is found in most oximeters and thus is readily available for routine use by anesthesiologists. Reduction of plethysmographic pulse wave amplitude (PPWA) has been proven to be a reliable method for detecting the IV injection of an epidural epinephrine-containing test dose under inhaled anesthesia in adults9 and children.10 This method, however, has not been tested during total IV anesthesia (TIVA). Additionally, PPWA was assessed in previous studies either visually or was obtained using sophisticated hardware and connections. Currently, a numerical value has been added to new pulse oximeters indicating the PPWA, termed the perfusion index (PI), to augment its clinical applicability.11

The aim of this study was to investigate whether changes in PI can be used as a new method for detecting intravascular injection of an epinephrine-containing test dose in adults under stable propofol-based anesthesia and to compare its reliability with the conventional hemodynamic criteria.


After local research committee approval and informed patient consent, 40 ASA I or II patients, scheduled for elective surgery under general TIVA were included in the study. Exclusion criteria were a history of smoking, diabetes mellitus, cardiovascular diseases, or use of medications affecting the cardiovascular system. Patients were randomly allocated using an online research randomizer ( into two groups (20 patients each) to receive IV either 3 mL of lidocaine 15 mg/mL with epinephrine 5 μg/mL or 3 mL of isotonic sodium chloride solution. Patients were premedicated with 10 mg diazepam orally 90 min preoperatively. Electrocardiographic heart rate (HR) and noninvasive oscillometric arterial blood pressure (BP) were measured using an S/5 anesthesia monitor (Datex-Ohmeda, Finland). The PI was monitored using a Masimo Radical SET (Masimo Corporation, Irvine, CA). For Masimo Radical SET, the PI upper and lower limits reported by the manufacturer are 0.02%–20.00%. The oximeter probe used to monitor the PI was attached to the middle fingertip of the hand contralateral to the site of BP monitoring and was wrapped in a towel to minimize heat loss and contamination by ambient light.

Anesthesia was induced with sufentanil 0.25 μg/kg administered over 1 min using a regular syringe pump and was maintained with a continuous sufentanil infusion of 0.0025 μg · kg−1 · min−1. Propofol was administered by a target-controlled infusion system (Vial Master TCI pump incorporating Diprifusor, Fresenius Vial SA, Brezins, France), in which an electric syringe is controlled by a computer incorporating a pharmacokinetic model. The model, which is based on the age and weight of the patient, continuously determines the infusion rate necessary to reach and maintain a given calculated plasma concentration of propofol.12 Propofol target-controlled infusion was started 1 min after opioid administration at an initial target plasma concentration of 3 μg/mL. Tracheal intubation was facilitated with rocuronium 0.6 mg/kg IV. Anesthesia was maintained with a target propofol concentration 2.5 μg/mL in addition to 60% air in oxygen. Patients' lungs were mechanically ventilated and minute volume was set to maintain end-tidal CO2 at 30–35 mm Hg. Fluid administration was standardized to 10 mL · kg−1 · h−1 of Ringer's lactate solution, and the ambient temperature was maintained at 25°C–26°C.

When hemodynamic variables and PI were stable for 5 min and at least 10 min had elapsed after anesthetic induction, patients were randomized to receive either 3 mL of isotonic saline (n = 20) or 3 mL of 1.5% lidocaine containing 15 μg of epinephrine IV (n = 20) as a simulated test dose via a peripheral IV catheter over 5 s, flushed with 10 mL of saline. After injection, BP cycling was set to every minute for 5 min. Data were collected at 20-s intervals for HR and PI, whereas for systolic BP (SBP), it was at 1-min intervals. In addition, maximal HR and PI responses were recorded. Anesthesia was conducted, and data were collected by the attending anesthesiologist who was blinded to the injected test dose. All measurements were made with the patient in the supine position before starting surgery.

Power analysis was based on a pilot study that resulted in a peak percent decrease in PI of 2% ± 2% after injection of saline and a 22% PI variability after injection of epinephrine-containing test dose. To detect a maximum PI difference of 25% from the preinjection values with Type I error of 0.05 and a Type II error of 0.2, 17 patients were required in each group. Positive HR, SBP, and PI responses to IV test dose were prospectively defined from previous reports9,13,14 as a HR increase of ≥10 bpm, a SBP increase of ≥15 mm Hg, and a PI decrease ≥10% after the simulated test dose administration. Sensitivity (true positives/[true positives + false negatives]), specificity (true negatives/[true negatives + false positives]), positive predictive values (true positives/ [true positives + false positives]), and negative predictive values (true negatives/[true negatives + false negatives]) were determined for HR, SBP, and PI variables. In addition, maximal HR, SBP, and PI responses were noted.

Data were tested for normal distribution using the Kolmogorov-Smirnov test. Differences between groups in demographic data, and baseline values of hemodynamic variables and PI were analyzed using unpaired t-test or χ2 test as appropriate. For comparison of changes in the criteria with time, data were first analyzed by repeated-measures analysis of variance, and differences were then calculated by post hoc testing (Newman–Keuls test). Analysis was performed using Statistica software version 7.0 for windows (Statsoft, Tulsa, OK). The 95% confidence interval (CI) for sensitivities, specificities, positive, and negative predictive values were calculated using the Wilson procedure as described by Newcombe.15 The Wilson procedure was also used for calculating the 95% confidence limits for the difference between sensitivities using different criteria.16 Data were presented as mean ± sd in the text and Table 1 and as mean ± 95% CI in Figure 1.

Table 1:
Patient Characteristics
Figure 1.:
Changes in heart rate (HR, bpm), systolic blood pressure (SBP, mm Hg), and percent perfusion index (PI), expressed as change from the preinjection value, after injection of the test dose containing 15 μg of epinephrine in total IV anesthetized adults (n = 20). Because HR, SBP, and PI were essentially unchanged after saline injection, these data are not presented. Vertical bars denote 0.95 confidence intervals. *A significant difference versus preinjection values (time 0).


There were no significant differences between groups with respect to age, weight, height, and gender distribution. There were also no significant differences in the preinduction HR, SBP, and PI. After induction of anesthesia and achievement of a steady anesthetic concentration, SBP and HR decreased and PI increased significantly from preinduction values. There were, however, no significant differences between the two groups regarding preinjection (baseline) data (Table 1).

IV injection of the epinephrine-containing solution produced significant increases in HR and SBP in the test dose group. Maximal increases in HR and SBP were 19 ± 8 bpm at 49 ± 25 s and 17 ± 7 mm Hg at 102 ± 34 s after test dose injections, respectively. Moreover, injecting the test dose resulted in an average maximum PI decrease by 65% ± 13% at 39 ± 15 s.

The HR showed a biphasic response after injection of the epinephrine-containing test dose. It first increased significantly after injection from 20 to 60 s, and then it decreased significantly from 140 s until the end of the study period. There were also significant decreases in PI from the preinjection values between 20 and 120 s in the test dose group (Fig. 1).

As shown in Table 2, using the PI criterion (positive if PI decreases ≥10% from the preinjection value), the sensitivity, specificity, positive, and negative predictive values were all 100% (CI = 84%–100%). On the contrary, two subjects who received an intravascular epinephrine-containing test dose were missed using the HR criterion, and one subject was missed using the SBP criterion resulting in sensitivities of 95% (CI = 76%–99%) and 90% (CI = 70%–99%), respectively. The lower and upper 95% confidence limits for the difference between PI and HR sensitivities were 0.0765 and 0.301, whereas the limits for the difference between PI and SBP were −0.1163 and 0.2361.

Table 2:
Sensitivity, Specificity, Positive, and Negative Predictive Values Based on Hemodynamic and Plethysmographic Pulse Wave Amplitude Criteria


The main finding in the present study was that percent PI response criterion (a decrease ≥10%) achieved 100% sensitivity and specificity in detecting the intravascular injection of an epinephrine-containing epidural test dose during stable TIVA in adult patients. On the other hand, neither the HR nor the indirectly measured SBP criteria were 100% reliable in detecting intravascular injection of the test dose in this population.

PI is the numerical value of the amplitude of the plethysmographic pulse wave that is displayed on many pulse oximeters. Digital representation of the PPWA may enhance wide clinical applicability. Using pulse oximetry, a variable amount of light is absorbed by pulsating arterial flow (AC) and a constant amount of light is absorbed by nonpulsating blood and tissue (DC). The pulsating signal indexed against nonpulsating signal and expressed as ratio is commonly referred to as the “perfusion index” = AC × 100/DC%.17 Changes in finger PI correspond to changes in the blood volume pulsations and depend on the distensibility of the vascular wall and the intravascular pulse pressure.18 Usually the effect of autonomic impulses upon distensibility is so strong that it predominates the opposite effect of pulse pressure. Decreases in PI resulting from pain and other stressful stimuli are due to vasoconstriction of the finger arterial bed rather than changes in the pulse pressure.19 As with endogenous catecholamines, drugs cause a predictable reduction in PI.20 In the present study, injection of the epinephrine-containing test dose resulted in significant decreases in PI from 20 to 120 s after injection. The strong maximal PI reduction after IV injection of epinephrine (65%) may indicate that a smaller dose of epinephrine may be sufficient to produce a 10% reduction.

Contrary to previous reports that showed that the HR response was 100% reliable in detecting intravascular injection during propofol anesthesia,13,21 in the present study, HR criterion was not 100% reliable in detecting intravascular injection of a simulated test dose containing epinephrine during TIVA. This could have been due to the decreased HR response to epinephrine in propofol and sufentanil-anesthetized patients.22 This combination has a strong sympathetic inhibitory effect.22 In addition, the greater number of male patients in our study may explain this difference. It has been demonstrated that sympathetic vascular regulation is predominant in men compared with a dominant parasympathetic influence on HR regulation in women.23

The sensitivity of SBP criterion in the present study was 90% in detection of IV epinephrine. Our finding is in contrast to previous studies that showed 100% sensitivity.13,21 However, these studies measured BP invasively via indwelling arterial catheters, a practice that was not justified in our patients. The use of intermittent and noninvasive measurement of BP may have decreased its reliability to detect intravascular injection, because the temporary SBP increases may be easily missed between cycles.7 Failure to demonstrate 100% efficacy of the HR and noninvasive SBP criteria as markers for confirming intravascular test dose injection during propofol TIVA represents a limitation of these conventional hemodynamic responses. On the other hand, PI provided a reliable indicator of intravascular injection of epinephrine under TIVA. All patients experienced a decrease ≥10% from baseline in response to direct IV injection of epinephrine, without false-positive responses after saline injection, resulting in 100% sensitivity, specificity, and positive and negative predictive values.

Several measures were taken in the present study to eliminate sources of error with finger plethysmographic monitoring.24 Anesthetic technique, drugs, and monitoring site were standardized. Patients were maintained normothermic and their hands covered with a dark blanket to avoid regional hypothermia and contamination with ambient light. All measurements were made in the supine position with the transducer at the level of the heart. Importantly, to avoid individual variations in absorption, scattering and reflection of the emitted light, only within-subject changes in the PI were reported as every patient served as his or her own control.

There are few limitations in the current work. First, our study only enrolled adult patients receiving propofol-sufentanil anesthesia and those receiving different anesthetics may have a different PI response and thus require further study of the reliability of PI criterion. Second, a smaller epinephrine dose than that used in our work may be inadvertently injected IV during clinical practice. Therefore, a dose-response study is required to determine the minimum effective dose of epinephrine required to elicit 100% consistency of the PI method. Third, although this study showed that conventional hemodynamic criteria were imperfect markers of test dose injection, and that PI was a reliable indicator, it failed to demonstrate that the PI criterion was more effective than HR or noninvasive SBP criteria. This is because our study was not powered to detect this difference (Type II error). Retrospective power analysis of our data revealed that a minimum of 74 and 152 patients would be required to provide a power >0.8 (P = 0.5) for detecting a statistically significant difference in sensitivities between the PI and SBP criteria and between PI and HR criteria, respectively.

In conclusion, PI is an effective indicator of intravascular injection of an epinephrine-containing epidural test dose under TIVA in adult patients. Further studies are warranted to determine whether this method is still valid under diverse TIVA techniques or when using smaller doses of epinephrine.


1. Dunet F, Pfister Ch, Deghmani M, Meunier Y, Demeilliers-Pfister G, Grise P. Clinical results of combined epidural and general anesthesia procedure in radical prostatectomy management. Can J Urol 2004;11:2200–4
2. Agarwal A, Pandey R, Dhiraaj S, Singh PK, Raza M, Pandey CK, Gupta D, Choudhury A, Singh U. The effect of epidural bupivacaine on induction and maintenance doses of propofol (evaluated by bispectral index) and maintenance doses of fentanyl and vecuronium. Anesth Analg 2004;99:1684–8
3. Lee PK, Kim JM. Lumbar epidural blocks: a case report of a life-threatening complication. Arch Phys Med Rehabil 2000; 81:1587–90
4. Soltesz EG, Pelt FV, Byrne JG. Emergent cardiopulmonary bypass for bupivacaine cardiotoxicity. J Cardiothorac Vasc Anesth 2003;17:357–8
5. Mulroy MF, Norris MC, Liu SS. Safety steps for epidural injection of local anesthetics: review of literature and recommendations. Anesth Analg 1997;85:1346–56
6. Guinard JP, Mulroy MF, Carpenter RL, Knopes KD. Test doses: optimal epinephrine content with and without beta-adrenergic blockade. Anesthesiology 1990;73:386–92
7. Liu SS, Randall L, Carpenter MD. Hemodynamic responses to intravascular injection of epinephrine-containing epidural test doses in adults during general anesthesia. Anesthesiology 1996;84:81–7
8. Mowafi HA. Digital skin blood flow as an indicator for intravascular injection of epinephrine-containing simulated epidural test dose in sevoflurane-anesthetized adults. Anesth Analg 2005;101:584–8
9. Mowafi HA. The efficacy of plethysmographic pulse wave amplitude as an indicator for intravascular injection of epinephrine-containing epidural test dose in anesthetized adults. Anesth Analg 2005;101:1506–11
10. Mowafi HA, Arab SA, Ismail SA, Al-Ghamdi AA, Al-Metwalli RR. Plethysmographic pulse wave amplitude is an effective indicator for intravascular injection of epinephrine-containing epidural test dose in sevoflurane-anesthetized pediatric patients. Anesth Analg 2008;107:1536–41
11. Uemura A, Yagihara M, Miyabe M. Pulse oximeter perfusion index as a predictor for the effect of pediatric epidural block. Anesthesiology 2006;105:A1354
12. Englbers F. Practical use of ‘Diprifusor' systems. Anaesthesia 1998;53(suppl 1):28–34
13. Takahashi S, Tanaka M, Toyooka H. Fentanyl pretreatment does not impair the reliability of an epinephrine-containing test dose during propofol-nitrous oxide anesthesia. Anesth Analg 1999; 89:743–7
14. Glantz SA. Confidence interval for the entire population. In: Glantz SA, ed. Primer of biostatistics. 3rd ed. New York: Mcgraw-Hill, 1992;212–7
15. Newcombe RG. Two-sided confidence intervals for the single proportion: comparison of seven methods. Stat Med 1998;17: 857–72
16. Newcombe RG. Interval estimation for the difference between independent proportions: comparison of eleven methods. Stat Med 1998;17:873–90
17. Lima A, Bakker J. Noninvasive monitoring of peripheral perfusion. Intensive Care Med 2005;31:1316–26
18. Goldman JM, Petterson MT, Kopotic RJ, Barker SJ. Masimo signal extraction pulse oximetry. J Clin Monit Comput 2000;16: 475–83
19. Dorlas JC, Nijboer JA. Photo-electric plethysmography as a monitoring device in anesthesia. Br J Anaesth 1985;57:524–30
20. Hoffman BB, Lefkowitz RJ. Catecholamines and sympathomimetic drugs. In: Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, Gilman AG, eds. Goodman and Gilman's, The pharmacological basis of therapeutics. New York: Pergamon Press, 1996:199–284
21. Takahashi S, Tanaka M, Toyooka H. The efficacy of hemodynamic and T-wave criteria for detecting intravascular injection of epinephrine test dose in propofol-anesthetized adults. Anesth Analg 2002;94:717–22
22. Schricker T, Carli F, Schreiber M, Wachter U, Geisser W, Lattermann R, Georgieff M. Propofol/sufentanil anesthesia suppresses the metabolic and endocrine response during, not after, lower abdominal surgery. Anesth Analg 2000;90:450–5
23. Evans JM, Ziegler MG, Patwardhan AR, Ott JB, Kim CS, Leonelli FM, Knapp FC. Gender differences in autonomic cardiovascular regulation: spectral, hormonal, and hemodynamic indexes. J Appl Physiol 2001;91:2611–8
24. Blanc VF, Haig M, Troli M, Sauvé B. Computerized photo-plethysmography of the finger. Can J Anaesth 1993;40:271–8
© 2009 International Anesthesia Research Society