To Study the Role of Perfusion Index as a Predictor of Hypotension during Spinal Anesthesia in Lower Segment Cesarean Section – A Prospective Observational Study

Abstract

Background and Aims: 

Spinal anesthesia has become the choice of neuraxial anesthesia for cesarean delivery. Hypotension, the most common side effect of spinal anesthesia is more in parturients. Perfusion index (PI) is one of the newest methods, used noninvasively for the prediction of hypotension postspinal anesthesia, evaluation of regional block success, and a tool for pain assessment. In our study, we have used PI for predicting the occurrence of hypotension postsubarachnoid block in elective lower segment cesarean section (LSCS).

Methods: 

In this prospective observational study, parturients posted for elective LSCS under spinal anesthesia were included. Spinal anesthesia was performed at L₃-L₄ interspace using 25G (gauge) Quincke needle, with 1.8 mL of 0.5% hyperbaric bupivacaine plus 20 μg fentanyl. Hypotension was defined as a decrease in systolic blood pressure >25% from the baseline.

Statistical Analysis Used: 

Statistical analysis was performed using the paired sample t-test, Chi-square test, and independent t-test.

Results: 

From our study, we had found that a positive correlation exists between the baseline PI of the patient and the occurrence of hypotension intraoperatively. A cutoff value of 1.75 (with P < 0.001) was found to be significant; above which the occurrence of hypotension was definite. The sensitivity of this cutoff value was 75% and specificity was 71% for this value.

Conclusion: 

Baseline perfusion of >1.75 can predict hypotension postspinal anesthesia in LSCS. The incidence of hypotension was about 93% in parturients whose baseline was above the cutoff value.

INTRODUCTION

Spinal anesthesia has emerged as the anesthesia of choice for lower segment cesarean section (LSCS) owing to the disadvantages faced with general anesthesia such as difficult airway and risk of aspiration.[¹] While spinal blockade offers it advantages, it comes with the most common risk of postspinal hypotension. In pregnancy, the increased sensitivity of the nerve fibers to local anesthetics causes rapid sympatholysis.[²] The aorto-caval compression added to this rapid effect aggravates the degree of hypotension.[³] Sympathetic activity dominance over the parasympathetic activity has been noted in pregnancy leading to greater peripheral vasodilatation and decreased venous return and preload.[⁴] Compensatory baroreceptors mechanisms can be inhibited at higher levels of block resulting in bradycardia and cardiac arrest.

Perfusion index (PI) is defined as “the ratio of pulsatile blood flow to the nonpulsatile blood flow.”[⁵] The formula for its calculation is as follows;

Where,

AC is the infrared signal's pulsating component and is the light absorbed by the pulsatile blood flow in the artery.

DC is the infrared signal's nonpulsating component, which is absorbed by the skin and other tissues.

It is a simple and noninvasive measure of the strength of blood flow reflecting the quality of perfusion at the measured site.[⁶] It has been used a predictor of vasopressor requirement during sepsis, hypotension during hemofiltration and intermittent hemodialysis, success of regional blocks in anesthesia and determinant of hypotension during LSCS.

In our study, the PI value has been used to predict the development of hypotension postspinal anesthesia in parturients scheduled for elective LSCS as the physiological changes in these patients make them more prone to the deleterious effects of hypotension and this would provide a basis for early intervention.

METHODS

The prospective observational study was conducted from the time period of September 2019 to March 2020 after obtaining approval from the Institutional Ethics Committee. It was a double-blinded study. The doctor administering spinal anesthesia, the observer of the values of perfusion index and episodes of hypotension and its values ; and the patient were all blinded in the study. Every participant in the study was explained about the nature of the study and informed and written consent was obtained for the same.

Inclusion criteria

• Parturients posted for elective LSCS
• Age between 20 and 35 years.

Exclusion criteria

Parturients with:

• placenta previa
• preeclampsia
• cardiovascular disease
• cerebrovascular disease
• gestational diabetes mellitus
• gestational age <36 or >41 weeks
• Body mass index ≥40
• Emergency LSCS
• Contraindication to spinal anesthesia.

The patients taken for the study were parturients between the age group of 20 and 35 years posted for elective cesarean section. Patients were explained about the study and informed written consent were obtained.

Fasting of 8 h for solids and 2 h for clear fluids was observed.

The patients were premedicated with tablet pantoprazole 40 mg night before surgery and at 6 a.m. on the morning of surgery and tablet metoclopramide 10 mg at 6 a.m. on the morning of surgery.

The patient were transferred to the operation theater, and the standard monitors were connected, i.e., electrocardiogram, noninvasive blood pressure, and pulseoximetry. The pulseoximetry probe was connected to the left index finger.

Intravenous access was established in the left upper limb. Each patient will be prehydrated with 500 mL of ringer lactate.

After prehydration, the baseline hemodynamic values and the baseline PI was recorded with the patient in the supine position by an anesthetist who was not be involved in the further intraoperative monitoring. The perfusion index was measured from the Philips M40 monitor for all the patients, by connecting its attached pulse oximeter to all the patients under the study. It was measured as a part of the variation in the plethysmographic waveform.

Spinal anesthesia was performed by an anesthetist blinded to the baseline PI values, using 25 G (gauge) Quincke's spinal needle, with the patient in the left lateral decubitus position. 2 mL of 0.5% hyperbaric bupivacaine with 20 micrograms of fentanyl was given at the L₃-L₄ or L₂-L₃ interspace.

The patient was turned back to the supine position and the sensory block was checked after 5 min with cold swab. A level of T₆ was taken as adequate to start the surgery.

The systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), and PI were recorded every 2 min up to 20 min and then every 5 min till the end of surgery. Intravenous (i.v.) fluids (crystalloids) were administered at the rate of 100 mL.10 min⁻¹.

Hypotension was defined as decrease in SBP >25% from the baseline value and was treated with i.v. bolus of 6 mg of injection ephedrine and 100 mL of i.v. fluids.

Following extraction of the baby, injection oxytocin 10 units was given as uterotonic at the rate of 200 mU.min⁻¹. Patients requiring additional uterotonics and blood loss more than 1000 mL were excluded from the study. Side effects such as nausea and vomiting were recorded.

Presentation of categorical and discrete data is in the form of tables and graphs are used for continuous data. Regression analyses were done.

An receiver operating curve (ROC) was obtained to analyze the baseline PI and was correlated with the episodes of hypotension 63 patients.

Parameters observed

• SBP
• DBP
• MAP
• PI.

The above parameters were observed every 2 min up to 20 min and then every 5 min up to 50 min.

RESULTS

In our study, a total number of 63 paturients were evaluated. Two paturients were excluded in the initial phase of the study due to inadequate spinal blockade level and one parturient was excluded as additional oxytocic was required. The demographic parameters of all the parturients were comparable. Majority of the patients were between the age group of 20–30 years (68.3%), and the remaining were in the age group of 30 − 40 years [Graph 1].

Graph 1. Age in years

The mean and standard deviation of weight (kg) is 78.554 ± 10.271. The median weight (kg) is 79, and the range is 46 (53–99). The inter-quartile range is 11 which we get from Q1 (75) and Q3 (86) [Graph 2].

Graph 2. Weight (kg)

The mean and standard deviation of height (cm) is 162.286 ± 5.265. The median height (cm) is 162 and the range is 25 (153–178). The interquartile range is 8 which we get from Q1 (158) and Q3 (166) [Graph 3].

Graph 3. Height (cm)

The mean weight and the height of all the patients were also comparable, with the mean weight as 78.554 kg [Table 1] and the mean height being 162.28 cm [Table 2].

Table 1: Weight (kg)

Table 2: Height (cm)

The receiver operator curve is a way of showing the connection between the sensitivity and specificity of a cutoff value. In our study, we had constructed the ROC to arrive at our cutoff value. The cutoff value is 1.75.

Advantage of perfusion index

Pulse oximetry is one the American Society of Anesthesiologists (ASA) basic monitors and is used for all the cases irrespective of the type of anesthesia. The plethysmographic waveform variations have been used recently to detect fluid responsiveness, nociception, regional block success, and peripheral perfusion of the patient. PI provides us with vital information of prediction of hypotension postspinal anesthesia using the variation in the plethysmographic waveform calculated value.

Disadvantages of PI are more or less similar to that of pulse oximetry, the difference being that serial measurements throughout the procedure and its correlation with changes in the blood pressure have not been studied.

The Chi-square test result shows that there is a significant relation between PI baseline value and the presence of hypotension (χ² = 13.339, P = 0.000).

After comparing the demographic parameters, the based on the baseline PI values of all the patients, a cutoff value for the baseline PI was obtained by using the ROC curve [Graph 4]. From the curve, we have obtained a value of 1.75 as the baseline PI which was found to be significant with a P value of 0.000 and area under the curve as 0.780 [Table 3]. Further using the Chi-square test, the correlation between the baseline PI and the incidence of hypotension intraoperatively was evaluated and a significant correlation was found between the baseline PI and the occurrence of hypotension (P = 0.000) [Table 4]. This cutoff value provided a sensitivity of 75% and specificity of 71% [Table 5].

Graph 4. Receiver operating curve for baseline perfusion index against incidence of hypotension

Table 3: Receiver operating curve analysis of perfusion index baseline with the presence of hypotension

Table 4: Sensitivity analysis of perfusion index baseline with the presence of hypotension

Table 5: Sensitivity and specificity

DISCUSSION

Pregnancy is a state of increased metabolic demand. The cardiovascular changes are of utmost importance as they associate with the changes induced by central neuraxial blockade. Hormonal effects produce peripheral vasodilatation with decrease in the systemic vascular resistance.[⁷] The gravid uterus poses a risk of vena-caval compression, thereby reducing the venous return and decreased cardiac output as the gestational week increases.[⁸]

Despite the differences, spinal anesthesia has always been the anesthesia of choice for LSCS. The sympathetic blockade caused by spinal anesthesia further aggravates the decrease in systemic vascular resistance and venous return leading to profound hypotension. The requirement of vasopressors is also increased due to adrenergic down regulation in pregnancy.[⁹]

PI, which is a noninvasive tool to determine the occurrence of hypotension, has been used in this study. Previously studies have been conducted to predict this aspect not only postspinal anesthesia but also in critical care medicine and in patients with severe sepsis. Mowafi et al.[¹⁰] had used PI as an indicator of intravascular injection of epinephrine in epidural test dose and had found it to be a reliable value. Similarly, Ginosar et al.[¹¹] had proved it to be an earlier and more sensitive indicator of epidural-induced sympathectomy compared to MAP. In a study by Kupeli et al.,[¹²] where they had used PI as a tool for assessment of pain in labor analgesia, they had concluded that the PI increased with the decrease in pain and this was due to the sympathetic block and resulting increase in blood flow leading to an increase in the perfusion, and hence, the PI. Klodell et al.[¹³] had successfully shown that PI can be used an effective tool for the detection of intraoperative thoracic sympathetic blockade. Specifically, a study on the prediction of hypotension with PI in LSCS was first conducted by Toyoma et al.[¹⁴] and they had found a positive correlation between the decrease in arterial blood pressure during spinal anesthesia and the baseline PI. An ROC curve concluded by the study had derived a cutoff value of 3.5 as the baseline PI above which the risk of hypotension was increased. Duggappa et al.[¹⁵] conducted a similar study, in which they had divided the participants into two groups as above and below 3.5 baseline PI. With an area under the curve value of 0.848, the study concluded a positive correlation between the baseline PI and development of hypotension. The sensitivity of the value 3.5 was 69.84% and specificity was 89.29% in their study. George et al.[¹⁶] also concluded a significant correlation between PI and the decrease in the systemic arterial pressure, but they had derived a baseline PI of 3.6 as their cutoff value with a sensitivity 80% and specificity of 60%. Varghese[¹⁷] conducted a similar study and concluded the same positive correlation result with an area under the curve of 0.911 and with the same baseline PI of 3.5 having a sensitivity of 86.6% and specificity of 93.33%. A more recent study by Mallawaarachchi et al.[¹⁸] had correlated the trend in PI and the degree of hypotension. They had found a significant correlation between the increase in the PI value and the incidence of hypotension and also concluded that the response to vasopressors could be quickly assessed by this value.

In our study, 63 patients were who completed the study were analyzed [Diagram 1]. As in the previous studies, we have found a positive correlation between the baseline PI and the occurrence of hypotension postspinal anesthesia and the value is significant (P = 0.000). We have determined a new cutoff value from our ROC curve and a value of 1.75 has been determined as the baseline cutoff above which a hypotension postspinal anesthesia can be expected. The sensitivity of this value was found to be 75% and specificity of 71%.

Diagram 1. Consort flow diagram

From our study, we would like to conclude that, although in previous studies, a different baseline value has been used as standard, in our study have achieved at our own value by performing an ROC analysis curve without using a baseline standard value.

However, in our study, we have not determined the relation between the PI and blood pressure throughout the procedure, that is, a serial measurement and correlation has not been established, making it a limitation of our study. Hence, in future studies, minute to minute relation between the PI and the changes in blood pressure can help to give a better correlation between the two values.

CONCLUSION

From our study, we like to conclude that a baseline PI value of 1.75 would be able to predict the occurrence of hypotension postspinal anesthesia in LSCS.

Financial support and sponsorship

This study was financially supported by Dr. D. Dhivya and Dr. Prasanth Kumar.

Conflicts of interest

There are no conflicts of interest.