Perfusion index and perfusion index ratio as predictive tools for block success: A prospective observational study on ultrasound-guided supraclavicular brachial plexus block

Abstract

Background:

The supraclavicular brachial plexus block is widely used for upper limb surgeries due to its effectiveness in providing anesthesia. Conventionally, success is gauged through sensory and motor assessments, which are not only subjective but also require active patient cooperation. This poses challenges for patients who are sedated or under general anesthesia. Additionally, repeated sensory testing can be uncomfortable. Therefore, an objective and noninvasive method to assess block success is needed. Perfusion index (PI), derived from pulse oximetry, reflects peripheral perfusion changes following sympathetic blockade and could serve as a useful surrogate marker. This study aimed to assess the utility of PI and the PI ratio as early indicators of successful block onset and to identify optimal threshold values correlating with effective anesthesia.

Methods:

Seventy patients undergoing elective upper limb orthopedic procedures received ultrasound-guided supraclavicular blocks. Data for PI were collected from the affected and unaffected limbs at the initial time point and again at 10, 15, and 20 minutes after the block was given. The PI ratio was derived by comparing values between limbs at each time point.

Results:

In successful blocks, the PI increased significantly in the affected limb. A PI > 2.94 showed 50% sensitivity and 91.67% specificity, while a PI ratio > 1.25 offered 50.78% sensitivity and 100% specificity at 10 minutes, confirmed by receiver operating characteristic (ROC) analysis.

Conclusion:

Both PI and PI ratio are effective in predicting block success, with the PI ratio proving more reliable, especially at the 10-minute mark.

Introduction

The supraclavicular brachial plexus block, first described by Kulenkampff in 1911, provides reliable anesthesia for the upper limb, and with the advent of ultrasound guidance, its safety and success rates have significantly improved, making it the preferred technique in current practice.[1,2]

Traditional methods to evaluate nerve block efficacy rely on assessing sensory and motor responses, which are subjective, time-consuming, and need patient co-operation. These techniques are not feasible in sedated individuals or those under general anesthesia, highlighting the need for a more objective and reliable method of block assessment.[3] Moreover, repeated pin-prick or cold sensation tests used for this evaluation can be unpleasant for patients and may lead to dissatisfaction or anxiety.[4] In operating room settings, relying on subjective assessments can delay surgical workflow and complicate anesthetic decision-making.[5]

Recent research has investigated different approaches to objectively evaluate the effectiveness of nerve blocks, which included evaluation of the physiologic changes such as altered blood flow, skin temperature and vasodilation. These were carried out by thermographic temperature measurement, laser Doppler perfusion imaging and skin electrical resistance.[6] All of these methods have shown promise in research settings, but they require specialized equipment and additional training, making them less feasible for routine use in most clinical setups. The need for sophisticated equipment to measure these changes and the time consumed made it difficult to employ them in routine practice. Perfusion index (PI) was reported to increase after peripheral nerve blocks like sciatic nerve block (animal study), stellate ganglion block in adults, and digital nerve block and caudal anesthesia in children.[7,8,9] These findings suggest that changes in peripheral circulation, especially vasodilation due to sympathetic blockade, can serve as useful indirect indicators of successful block onset. The PI appears to be a simple, practical solution to this clinical challenge.

The PI, obtained noninvasively from pulse oximetry, represents the ratio of pulsatile arterial blood flow to nonpulsatile venous and tissue components, providing an objective measure of peripheral perfusion. Its simplicity and availability on standard monitors have led to increasing interest in its application for regional anesthesia. Recent clinical studies have shown that PI rises significantly after successful peripheral nerve blocks, supporting its role as an early indicator of block onset.[10,11,12] Furthermore, prospective evaluations have reported that both absolute PI values and PI ratios can reliably predict block success with high diagnostic accuracy.[13] These findings suggest that PI may serve as a practical, objective alternative to traditional sensory or motor testing in the intraoperative setting.

This study aimed to evaluate the role of PI and PI ratio as objective indicators of successful supraclavicular brachial plexus block. Specifically, we assessed changes in PI values and PI ratios at serial time points after block administration and compared them with conventional sensory and motor assessments. Our hypothesis was that both PI and PI ratio would increase significantly in successful blocks, and that a defined threshold could reliably predict block success within 10 minutes.

Methods

This study was conducted at a quaternary care center in urban India, in the Department of Anaesthesiology, from February 2021 to October 2022, using a prospective observational design. The study consisted of 80 adult patients who were slated to undergo elective orthopedic surgery on the upper limb [Figure 1]. Patients aged 18-60 years, of either sex, with American Society of Anaesthesiologists (ASA) Physical Status I or II, scheduled for elective upper limb orthopedic surgery were included. Exclusion criteria comprised patient refusal, ASA physical status III or IV, known cardiovascular or cerebrovascular disease, bleeding diathesis, local infection at the block site, BMI ≥ 40 kg/m², history of seizures, allergy to study drugs, peripheral vascular disease, pre-existing neurovascular injury to the upper limb, and diabetes mellitus with peripheral neuropathy. Institutional Ethics Committee (IEC) approval was obtained (MSRMC/EC/PG-72/01-2021) dated 28/01/2021, and written informed consent was obtained from all participants. This study was prospectively registered with the Clinical Trials Registry of India (CTRI) under registration number CTRI/2021/03/032380, registered on 30/03/2021. Recruitment had started in February 2021, and registration was therefore completed after participants had already been enrolled.

Figure 1.

CONSORT flow diagram of patient enrollment, exclusion, and final analysis

Procedure

Upon entering the operating room, patients were monitored using electrocardiography (ECG), automated noninvasive blood pressure monitoring, heart rate monitoring, and pulse oximetry. The premedication regimen consisted of administering IV Oondansetron at a dosage of 4 mg and midazolam at a dosage of 0.03 mg/kg of body weight. An anesthesiologist, who had at least 1 year of experience in ultrasound-guided peripheral nerve blocks, performed the supraclavicular brachial plexus block. Patients were placed in a supine position with their head turned towards the opposite side. The administration of the block was performed with great care to maintain a sterile environment. A linear ultrasound probe (8-12 MHz; GE Logitech Venue 40) enclosed in a sterile covering was used. The procedure involved inserting a 22G 50 mm Stimuplex needle in a straight line (long axis) under ultrasound guidance. The brachial plexus trunks/divisions were observed as a dense cluster of hypoechoic, circular, or oval nerves positioned to the side and on the surface of the subclavian artery, and above the first rib. An initial infiltration using 1 ml of 1% lignocaine was performed at the skin. This was followed by incremental administration (in 5 ml portions) of a 30 ml local anesthetic mixture comprising 20 ml of 0.75% ropivacaine, 10 ml of 2% lignocaine, and 4 mg of dexamethasone. The solution was carefully deposited around the neural structures ensuring circumferential spread within the plexus sheath and visualization of the “corner pocket” spread.[14]

Sensory and motor block assessments were performed at 5-minute intervals until 20 minutes postinjection. The dermatomes of the median, ulnar, radial, and musculocutaneous nerves were examined with a pinprick stimulus to evaluate sensory blockade. Motor block was assessed by the patient’s ability to perform elbow flexion and hand movements against gravity. Block success was defined as complete sensory loss across the C5–T1 dermatomes, while the requirement for general anesthesia owing to insufficient block was classified as a unsuccessful block. Patients who required conversion to general anesthesia or had failed blocks were not excluded from the study, but their data were analyzed separately to allow evaluation of PI values in both successful and unsuccessful blocks. PI values were recorded using Datex-Ohmeda pulse oximeter probes placed on the index fingers of both the blocked and unblocked limbs. Baseline PI was measured before block administration, and subsequent readings were recorded at 10, 15, and 20 minutes postinjection. For analysis, the PI from the blocked limb was compared with the contralateral limb. The PI ratio was defined as the ratio of the PI at 10 minutes postblock to the baseline PI value. All patients were monitored for potential complications including bradycardia (heart rate < 50 bpm), hypotension (≥30% decrease from baseline blood pressure), nausea, vomiting, and hypoxemia (oxygen saturation < 90%). Appropriate interventions were undertaken as required.

Statistical analysis

Data analysis was performed using SPSS Statistics for Windows, version 22.0 and R statistical software version 3.2.2. Descriptive statistics for continuous variables were expressed as mean ± standard deviation (SD), along with their respective minimum and maximum values. For variables not normally distributed, data were summarized using median and interquartile range (IQR). The distribution of categorical variables was described using percentages and frequencies. A P value of <0.05 was considered statistically significant. Comparisons between two independent groups were performed using the two-tailed independent Student’s t-test for normally distributed data. Paired t-tests were applied to evaluate within-group differences for paired observations. For variables not meeting the assumption of normality, analysis was performed using the Wilcoxon signed-rank test. Receiver operating characteristic (ROC) curve analysis was employed to assess the discriminative ability of the PI as a predictor of successful supraclavicular brachial plexus block. The diagnostic accuracy of PI was quantified using the area under the ROC curve (AUC). The AUC reflects the overall ability of the parameter to correctly classify cases. An AUC of 0.5 indicates no discriminative ability, whereas values closer to 1.0 reflect greater diagnostic accuracy. Optimal cutoff values for the PI and PI ratio were determined using the You den index (sensitivity + specificity – 1) derived from the ROC curve analysis.

Sample size estimation was carried out using MedCalc Statistical Software. The calculation was based on findings from a previous study on supraclavicular brachial plexus block, which reported an area under the curve (AUC) of approximately 0.80 for perfusion index in predicting block success.[15] For the present study, we assumed an expected AUC of 0.70 compared with a null hypothesis AUC of 0.50. The choice of 0.70 was made as a conservative estimate, reflecting both clinical relevance and values reported in earlier literature.[13] With a two-sided significance level of 0.05 and 70% power, the minimum required sample size was 70 patients. A power of 70% was selected in view of the practical constraints of case availability during the study period, while still allowing detection of a clinically meaningful effect. To compensate for potential exclusions or dropouts, the target enrolment was increased by approximately 10–15%, yielding a final planned sample size of 80 patients.

Results

A total of 80 patients were enrolled in the study. Of these, 10 were excluded (6 due to incomplete PI recordings, 4 due to excessive movements and inability to cooperate). [Figure 1] The remaining 70 patients were included in the final analysis.

Baseline demographic and clinical characteristics of the two groups are summarized in Table 1. No significant differences were observed except for body mass index (BMI), which was higher in the successful block group (P = 0.022). The distribution of clinical diagnoses among the 70 patients is presented in Table 2. Distal radius fracture was the most common indication, while other upper limb fractures and soft tissue injuries were less frequent.

Table 1.

Comparison of clinical variables in successful and nonsuccessful limb blocks

Variables	Successful	Unsuccessful	P
Age in years	35.13±12.76	40.83±11.92	0.296
Height in cm	164.73±6.41	165±4.73	0.922
Weight in kg	65.3±8.49	60.17±10.32	0.169
BMI (kg/m2)	23.96±1.91	21.99±2.68	0.022*

Table 2.

Distribution of clinical diagnoses among patients undergoing supraclavicular brachial plexus block

Diagnosis	No. of Patients (n=70)	%
Distal radius fracture	19	27.1
Distal humerus fracture	8	11.4
Hand crush injury	7	10.0
Radius fracture	7	10.0
Montaggia fracture	3	4.3
Digital humerus fracture	3	4.3
Barton’s fracture	2	2.9
Forearm both bone fracture	2	2.9
Distal humerus shaft fracture	1	1.4
Epicondyle fracture	1	1.4
Fingers crush injury	1	1.4
Hand tendon injury	1	1.4
Little finger fracture	1	1.4
Medial epicondyle fracture	1	1.4
Midshaft of humerus fracture	1	1.4
Proximal forearm swelling	1	1.4
Radius + ulna fracture	1	1.4
Left ulna fracture	1	1.4
Midshaft humerus fracture	1	1.4
Hand infected wound	1	1.4
Metacarpal dislocation	1	1.4
Metacarpel K-wiring	1	1.4
Radius and ulna fracture	1	1.4
Ring finger bone exposed	1	1.4
Thumb flap	1	1.4
Ulna fracture	1	1.4

On clinical evaluation, most patients achieved complete sensory and motor block within 20 minutes, while a small proportion were classified as failures. The rise in PI and PI ratio was seen only in patients with clinically successful blocks, whereas values remained unchanged in those with incomplete or failed blocks, as shown in Table 3. This parallel trend supports the use of PI as an objective correlate of clinical block assessment.

Table 3.

Clinical block assessment and PI findings

Parameter	Successful block (n=64)	Failed block (n=6)
Complete sensory + motor block	64 (100%)	0
Time to complete sensory block – at 10 min	28 (43.8%)	0
Time to complete sensory block – at 15 min	22 (34.4%)	0
Time to complete sensory block – at 20 min	14 (21.8%)	0
PI at baseline	1.15±0.5	1.82±1.72
PI at 10 min	6.22±2.0	1.23±1.08
PI ratio	5.92±2.13	0.74±0.41

In the successful block group, PI values in the blocked limb rose sharply from baseline (1.15 ± 0.5) to 10 minutes (6.22 ± 2) and remained consistently elevated at 15 and 20 minutes. In contrast, the unblocked limb values stayed near baseline throughout. The PI ratio in the successful group showed a marked increase (5.92 ± 2.13). In the unsuccessful block group (n = 6), blocked limb PI values showed no appreciable rise from baseline (1.82 ± 1.72) over 10, 15, or 20 minutes, remaining comparable to the unblocked limb as shown in Figure 2. The PI ratio also failed to increase (0.74 ± 0.41).

Figure 2.

Trends of Perfusion Index (PI) values in blocked and unblocked limbs among successful and unsuccessful blocks. Data are presented as mean values (lines) with standard deviation (SD) represented by shaded areas

ROC curve analysis was performed to evaluate the diagnostic performance of PI values in predicting successful supraclavicular brachial plexus block [Figure 3] At 10 minutes postblock, a PI value cutoff of >2.94 (derived using You-den index) demonstrated a sensitivity of 91.67% and specificity of 98.46%, with a positive predictive value (PPV) of 50.00% and a negative predictive value (NPV) of 64.71%. The AUC was 0.707 (standard error [SE]: 0.0723), indicating fair diagnostic accuracy (P = 0.004). For the PI ratio, a cutoff value of >1.25 yielded both 100% sensitivity and specificity, with a PPV of 50.78% and an NPV of 67.01%. The AUC for PI ratio was 0.828 (SE: 0.064), demonstrating good diagnostic accuracy with strong statistical significance (P < 0.001), as shown in Figure 4. These findings suggest that both absolute PI values at 10 minutes and PI ratio are effective predictors of successful nerve block, with the PI ratio offering superior diagnostic performance.

Figure 3.

Receiver operating characteristic (ROC) curve for perfusion index (PI) at 10 minutes in predicting successful supraclavicular brachial plexus block. The area under the curve (AUC) was 0.707 (SE = 0.0723, P = 0.004), with an optimal cutoff value of >2.94

Figure 4.

Receiver operating characteristic (ROC) curve for PI ratio (blocked/unblocked limb) in predicting successful supraclavicular brachial plexus block. The AUC was 0.828 (SE = 0.064, P < 0.001), with an optimal cutoff value of >1.25

ROC curve analysis showed that PI at 10 minutes had a discriminatory ability of 70.7% (AUC = 0.707) for identifying a successful block [Table 4]. In comparison, a PI ratio >1.0 demonstrated higher accuracy with an AUC of 0.828, sensitivity of 82.03%, and specificity of 66.67%. However, at a higher cut-off of >1.25, the PI ratio showed reduced sensitivity (50.78%) but perfect specificity (100%). No adverse events or block-related complications were observed in any patient.

Table 4.

Receiver operating characteristic (ROC) analysis of perfusion index (PI) and PI ratio in predicting successful supraclavicular brachial plexus block

Variables	ROC results to predict successful limb				Cut-of f	AUROC	SE	P
	Sensitivity	Specificity	PPV	NPV
PI at 10 min	50.00	91.67	98.46	64.71	>2.94	0.707	0.0723	0.004*
PI ratio	50.78	100.00	100.00	67.01	>1.25	0.828	0.064	<0.001**

Discussion

The present study evaluated the role of PI and PI ratio in predicting the success of supraclavicular brachial plexus blocks, with a focus on early, noninvasive detection. Our findings support the use of the PI ratio as a reliable parameter, offering real-time insight into the efficacy of regional anesthesia. The results were interpreted in light of demographic, clinical, and physiological factors, as well as compared with similar studies in the literature.[13] This study differs from earlier reports by evaluating both PI and PI ratio side by side, with direct comparison between successful and failed blocks. Many previous works focused only on absolute PI values, whereas our use of the ratio helped reduce baseline variability and gave a clearer picture of perfusion changes over time. Including unsuccessful blocks as a control group also adds strength, since it highlights the contrast in perfusion patterns more distinctly than studies that examined only successful blocks.

The primary objective parameter under investigation, the PI, showed significant changes postblock in the successful group. A marked increase in PI was observed in the blocked limb as early as 10 minutes following the block, consistent with the physiological mechanism of sympathetic blockade leading to vasodilation. The unblocked limb and the failed blocks did not demonstrate any significant PI changes over time. This localized and time-bound rise in perfusion supports the use of PI as a surrogate marker for sympathetic block success. In order to mitigate interindividual variability in baseline perfusion index values, the study utilized the PI ratio, calculated by normalizing the PI at each time point to the respective baseline value. This approach enabled a more precise evaluation of relative changes in perfusion and improved the clinical relevance of the results. Our results are in concordance with the observations of Lima et al., who highlighted the considerable baseline variability of the perfusion index and advocated for the PI ratio as a more reliable parameter.[16]

In the present study, a PI value exceeding 2.94 at 10 minutes and a PI ratio greater than 1.25 were identified as optimal cut-off points. These thresholds closely match those reported by Abdelnasser et al. and Galvin et al., despite slight variations in technique and population characteristics.[15,17] Such consistency suggests that a standardized PI ratio cut-off could be applied across different block types and clinical settings. Our results also aligned with those of Shah et al. and Paul et al., who demonstrated a similar trend of rising PI in successful blocks. For instance, Paul et al. documented PI increasing from 1.97 to 7.46 postblock, while we noted a rise from 1.15 to 6.22. This trend reinforces the observation that sympathetic blockade reliably induces detectable changes in perfusion, supporting the utility of both PI and PI ratio as dependable indicators for assessing block efficacy.

ROC analysis showed that the PI at 10 minutes had an AUC of 0.707, while the PI ratio demonstrated better discriminative ability with an AUC of 0.828. The optimal cutoff was determined using the You den index, which identified a PI ratio >1.25 at 10 minutes as the best threshold for predicting block success. At this value, the PI ratio achieved 100% specificity but only moderate sensitivity (50.8%). These findings indicate that the PI ratio functions more reliably as a confirmatory tool rather than a screening measure, and its use may be most appropriate when combined with clinical assessment.

The overall sensitivity of the PI ratio was modest, highlighting the need to use it in conjunction with clinical indicators. A limitation of this study is that trial registration was done after participant enrollment had already started, which could lead to reporting or selection bias. However, the study protocol was prepared in advance and followed throughout. Also, the sample size was limited and restricted to healthy individuals, which may affect generalizability. Additionally, measurements were taken at fixed intervals up to 20 minutes, potentially missing earlier or delayed responses in some cases.

Conclusion

The PI ratio at 10 minutes is a simple, noninvasive tool that reliably predicts the success of supraclavicular brachial plexus block. A ratio >1.25 strongly correlates with block success. Clinically, this offers an objective and early method to confirm block efficacy, reducing unnecessary delays or conversion to general anesthesia.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.