Role of perfusion index in early detection of hypotension following spinal anesthesia for cesarean section

ABSTRACT

Background

The most common drawback of spinal anesthesia (SA) is hypotension, which has adverse effects on both mother and fetus; consequently, early detection and management of spinal hypotension have been considered a very important topic in obstetric anesthesia. Additionally, the capability to ultimately diagnose hypotension could permit initiation of adequate treatment. The standard method to estimate blood pressure during cesarean section is noninvasive blood pressure measurement. However, periods of hypotension could be missed within a reasonable time frame.

Methods

Perfusion index (PI) and blood pressure were measured in 100 pregnant women during elective cesarean section under SA.

Results

PI had a significant negative correlation with the change occurred in systolic blood pressure (SBP) (r = −.432, p < .0001); by using univariate linear regression analysis, the PI explains 20.2% of variability in SBP as PI significantly predicts the change occurred in SBP (F = 24.8, p < .0001), as each one unit increase in PI, accompanied by a significant decrease in SBP by 2.089 (mm/hg), as (b, t, p) (−2.089, −4.987, .0001), respectively. There was a significant difference in SBP and PI before and after giving oxytocin bolus [(z = −1.9, P = .05) and (z = −4.3, P < .0001), respectively]. By comparing PI before vs. after taking ephedrine bolus, there was a significant decrease in median of PI [median of PI before 10 (5.5–15) vs. median of PI after 7.5 (6–15.5) (z = −6.6, P < .0001)].

Discussion

After SA with sympathetic blockade, peripheral vascular tone declined and therefore PI increased.

Conclusion

PI could anticipate the incidence of hypotension during cesarean delivery with SA. Ephedrine effect could be evaluated rapidly by the alteration in the PI so that further need can be judged. Oxytocin has a significant impact on vascular tone with 5 international unit bolus, although it does not have a significant effect on blood pressure.

KEYWORDS:

• Anesthesia
• hypotension
• obstetric
• perfusion index
• pulse oximeter

1. Introduction

The most common technique of anesthesia in parturient for cesarean delivery is spinal anesthesia (SA) [1]. The rate of cesarean section imposed under SA has marked expansion in the last two decades [2].

The most common drawback of SA is hypotension, which has adverse effects on maternal and baby sequel. Consequently, early detection and management of spinal hypotension have been considered a very important topic in obstetric anesthesia. Additionally, the standard method to estimate blood pressure during cesarean section is noninvasive blood pressure (NIBP) measurement. However, periods of hypotension could be missed within a reasonable time frame.

Perfusion index (PI) is a noninvasive continuous estimation of peripheral perfusion based on pulse oximeter. It is the proportion of pulsatile and fixed non-pulsatile flow of blood in peripheral tissue like a fingertip which adds more convenient details in different clinical conditions [3]. PI could assess peripheral perfusion that can be related to vascular tone, which has a major effect on hypotension [4]. Previous research by Toyamo et al. proposed that, after SA for cesarean delivery, PI >3.5 could predict the incidence of hypotension [5].

It was hypothesized that changes in PI after SA could help in early detection of hypotensive episodes. It also detects response to ephedrine bolus after hypotensive episode; therefore, it can help in early detection and management.

2. Aim of the work

The primary aim of this study was to establish whether the trend of the perfusion index detects hypotension before the changes in NIBP values.

The secondary aim was to detect oxytocin bolus impact on the PI and the effectiveness of ephedrine dose during hypotensive episodes on PI and whether PI can assess the response to ephedrine [6,7].

3. Methods

In El Shatby Hospital, Alexandria, 100 elective cesarean deliveries were done in prospective observational study after taking a consent from pregnant female from November 2021 till November 2022 after acceptance of ethical committee in monthly meeting of 20 January 2022 with serial number 0305448.

4. Indications for elective cesarean section

• Breech presentation (at term) – planned cesarean sections for breech presentation at term have increased significantly since the “Term Breech Trial” [Lancet, 2000].

• Other malpresentations – e.g., unstable lie (a presentation that fluctuates from oblique, cephalic, transverse, etc.), transverse lie, or oblique lie.

• Maternal request – this covers a variety of reasons from previous traumatic birth to “maternal choice”.

Inclusion criteria were ASA (American Society of Anesthesiologists) II elective cesarean section of a single baby and 39-weak gestational ages or above.

Exclusion criteria were contraindication to SA, allergy to medication used, obese with BMI ≥35 kg/m², and height out of the range from 150 cm to 180 cm.

Preoperative pregnant patients were assessed carefully. Sodium citrate (30 ml) was given as prophylactic antacid in operating room after 5 minutes of rest, with the table tilt of 15 to 30 degree left lateral. Large cannula 18G was put after lidocaine 2%, a local anesthesia. Basic monitors were attached. Basal reading of heart rate, blood pressure, and PI was done by averaging of three consecutive readings. Heart rate and PI were monitored using pulse oximeter, and blood pressure (systolic mean diastolic) was monitored using automated noninvasive device (Biolight M69, Biolight Meditech®, Zhuhai, China).

 In the sitting position spinal anaesthesia was delivered under full sterilization. After skin local anaesthetic injection with lidocaine 2% 2 ml at the L3-L4 or L4-L5 vertebral interspace. At L3-L4 or L4-L5 intervertebral space, local anesthetic, 2.2 ml of 0.5% hyperbaric bupivacaine, with 25 μg (0.5 ml) fentanyl was given (total volume 2.7 ml) with Quincke spinal needle after confirmation of free flow of cerebrospinal fluid, and then the basal posture was restored.

As regards fluids, Ringer’s lactate coload was started with the injection of local anesthetic for the first 1 L. The bag of fluid was suspended at 1 m height above operating table midpoint. Maintenance fluids (10 ml/kg/hour) were given till the end of operation [8,9].

Block adequacy was assessed by anesthesiologist, and surgery could be started. Ice swap was used for block height and deemed adequate if T4 level. Bromage score was employed for motor block, and grade IV must be reached. In case of inappropriate sensory or motor level, the patient was excluded. Oxytocin 5 international unit IU as intravenous bolus was given after delivery of the baby then 2 international units. Hemodynamic monitors – (systolic, diastolic, and mean) blood pressure, heart rate, and PI – were measured every 1 minute till delivery and then every 3 minutes till the end of surgery.

The definition of hypotension was a decrease from the basal systolic blood pressure (SBP) by >20%, which is the most common definition in research study. Treatment with 10 mg ephedrine was given intravenously when mean blood pressure was below 65 mmHg [10].

Atropine 0.5 mg was given intravenously when heart rate was <60.

Oxygen was delivered by face mask to maintain oxygen saturation more than 95%.

5. Sample size calculation

Based on this study [11], a sample size of 100 achieves 80% power that detects effect size of 30% with a target significance level at 5%. NCSS 2004 and PASS 2000 program were used for sample size calculation.

6. Statistical analysis

The Statistical Package for the Social Sciences version 25 was used for data coding.

According to the distribution of variables using Kolmogorov–Smirnov test, data were summarized by median as measures of central tendency and range as measures of dispersion for quantitative not normally distributed, while for quantitative normally distributed variables, we summarized it using mean and standard deviation.

The strength and correlation course of PI and hypotension was estimated by Spearman correlation.

Univariate linear regression analysis was used to predict the sequel of PI on SBP. Wilcoxon test was used to detect significant difference in median quantitative variables among SBP, mean arterial blood pressure (MABP), and PI before and after giving oxytocin bolus.

7. Results

Study was done on 100 pregnant females; their mean age was 28.34 ± 5.1 years (ranged from 19 to 39 years), and the mean of their BMI was 28.13 ± 2.5 (ranged from 23 to 34). As regards gravida G, G2 constituted the majority (30%), followed by G3 (23%), then G1 and G4 were the same percent (20%), and G5 (8%) was the least. As regards parity P, P1 constituted the majority (34%), followed by P0 (26%), then P2 (19%), and P4 (9%) was the least. As regards gestational age GA, GA39 constituted the majority (42%), followed by GA40 (39%), and GA41 (19%) was the least. Basal mean PI was 3.7 ± 1.08 (ranged from 2 to 6.4).

Gravida ranged from 1 (20 cases), 2 (30 cases), 3 (23 cases), 4 (20 cases), and 5 (8 cases); parity ranged from 0 (26 cases), 1 (34 cases), 2 (19 cases), 3 (12 cases), and 4 (9 cases); and gestational age ranged from 39 weeks (42 cases), 40 weeks (39 cases), and 41 weeks (19 cases). Basal mean PI was 3.7 (ranged from 2 to 6.4). Figure 1 illustrates mean changes that occurred in PI, MABP, and SBP starting from baseline to 60 minutes.

Figure 1. Line graph describes mean changes that occurred in PI, MABP, and SBP starting from baseline to 60 minutes of cesarean section.

Spearman correlation revealed that the change that occurred in PI had a significant negative correlation with the change that occurred in SBP (r = −.432, p < .0001) (Table 1 and Figure 2).

Figure 2. Correlation between change in PI and change in systolic blood pressure during spinal anesthesia for cesarean delivery.

Table 1. Correlations between changes that occurred in the systolic blood pressure and in perfusion index.

		Correlations
		Systolic BP	Perfusion index
Systolic BP	R	1	−.439*
	P		<.0001*
	No.		100

Note: R; Spearman correlation.

*Statistically significant.

Using univariate linear regression analysis, the model including PI explains 20.2% of variability in SBP as PI significantly predicts the change occurred in SBP (F = 24.8, p < .0001), as each one unit increase in PI, accompanied by a significant decrease in SBP by 2.089 (mm/hg), as (b, t, p) (−2.089, −4.987, .0001), respectively (Table 2).

Table 2. Univariate linear regression analysis to predict the effect of PI on SBP.

	Unstandardized coefficients		Standardized coefficients	t	p	95% Confidence interval
	B	Std. error	Beta
(Constant)	−10.646	.883	-	−12.057	<.0001*	(−12.39 to −8.89)
PI	−2.089	.419	−.450	−4.987	<.0001*	(−2.920 to −1.258)

Note: Dependent variable: SBP.

Table 3 and Figure 3 show diagnostic ability of PI in the prediction of hypotension that occurred during cesarean section as change in PI can *; statistically significantly detect hypotensive cases with AUC was .92 (0.859 to 0.970) (P < 0.0001), the cutoff point of PI change with highest sensitivity (95% CI) and specificity (95% CI) [(>2.127, 73.7 (56.9–86.6) and 98.7 (94.2–100), respectively].

Figure 3. Diagnostic ability of PI in prediction of hypotension during cesarean section.

Table 3. Diagnostic ability of PI in prediction of hypotension during cesarean section.

NPV (95% CI)	PPV (95% CI)	Specificity (95% CI)	Sensitivity (95% CI)	Cutoff point	P value	AUC (95% CI)
86.1 (75.9–93.1)	100 (87.2–100)	98.7 (94.2–100)	73.7 (56.9–86.6)	>2.127	<.0001*	.92 (0.859–0.970)	PI

There was a significant difference in SBP and PI before and after giving oxytocin bolus [(z = −1.9, P = .05) and (z = −4.3, P < .0001), respectively], while there was no significance detected in MABP (z = −.86, P = .38), as shown in Table 4.

Table 4. Relations between the blood pressure variables and perfusion index following oxytocin bolus.

	Before oxytocin bolus (No. = 100)	After oxytocin bolus (No. = 100)	Test of significance (P)
SBP	103 (88–115)	105 (93–114)	(z = −1.9, P = .05*)
MABP	67.33 (59–78)	69 (65–75)	(z = −.86, P = .38)
PI	4.5 (2.8–12.4)	5 (2.9–11)	(z = −4.3, P < .0001*)

Note: Data described by median (Min-Max).

z; Wilcoxon test.

*Statistically significant.

During operation, ephedrine bolus was given 83 times to patients as taken once in 10 cases, twice in 11 cases, thrice in 7 cases, 4 times in 5 cases, and 5 times in only 2 cases.

By comparing PI before vs. after taking ephedrine bolus, there was a significant decrease in median of PI [median of PI before 10 (5.5–15) vs .median of PI after 7.5 (6–15.5) (z = −6.6, P < .0001)] (Table 5 and Figure 4).

Figure 4. Relations between perfusion indexes before ephedrine bolus vs. after ephedrine bolus.

Table 5. Relations between perfusion indexes before ephedrine bolus vs after ephedrine bolus.

	Before ephedrine bolus (No. = 83)	After ephedrine bolus (No. = 83)	Test of significance (P)
PI	10 (5.5–15)	7.5 (6–15.5)	(z = −6.6, P < .0001*)

8. Discussion

Steady hemodynamic parameters result in better final outcome. The most frequent drawback after SA is hypotension. Intraoperative intermittent NIBP and heart rate are the main methods to detect hypotension. Invasive blood pressure or advanced monitors are needed to have continuous detection of blood pressure [4].

PI can be measured from peripheral pulse oximeter so it can be used practically as a noninvasive test [3]. During pregnancy, blood volume and cardiac output expand. On the other hand, systemic vascular resistance declines, which matches with increase in PI due to vasodilatation [3]. After SA, there is high rise in sympathetic block level than sensory block; peripheral vascular tone is declined, and there is more hypotensive effect [12].

Spearman correlation revealed that the PI had a significant negative correlation with SBP. With multiple regression model, PI significantly predicts the change occurred in SBP, as each one unit increase in PI, accompanied by a significant decrease in SBP by 2.089 (mm/hg). This can be compared with the study of Roshana Prasad et al. which found that PI was a quick predictor of the incidence of hypotension during cesarean delivery [13].

In previous studies, PI has been reported to increase following the arterial vasodilation caused by sympathetic block [11]. Furthermore, PI measurement has been shown to be effective in evaluating the success of many peripheral nerve blocks, such as axillary block, infraclavicular block, sciatic block, stellate ganglion block, and interscalene block [11,14,15]. Ginosar et al. found that, after epidural anesthesia, PI increase was a reliable marker of sympathectomy [15].

9. Conclusion

PI could anticipate the incidence of hypotension during cesarean delivery with SA. Ephedrine effect could be evaluated rapidly by the alteration in the PI so that further need can be judged. Oxytocin has a significant impact on vascular tone with 5 international unit bolus, although it does not have a significant effect on blood pressure.

10. Limitations

Weak points in this study were found. First of all was the patient movement; therefore, PI should be measured with utmost care to avoid that. Level of anxiety must be asked as PI may decrease with anxiety.

Abbreviations

IV	=	intravenous
NIBP	=	noninvasive blood pressure
PI	=	perfusion index
SBP	=	systolic blood pressure
SA	=	spinal anesthesia

Consent

Trial registration number 0305448.

Acknowledgments

Thanks to all my family who supported me.

Disclosure statement

The authors report that there are no competing interests to declare.

The study protocol was approved by Alexandria University ethics committee. It was prospective observational study.

Additional information

Funding

Self fund.