Prevention of Pain on Propofol Injection: A Comparison of Pretreatment with 5-Hydroxytryptamine 3 Antagonists Granisetron, Ondansetron, and Ramosetron

Abstract

Background and Aims:

Propofol is commonly used for the induction and maintenance of anesthesia. However, approximately 28%–90% of patients experience pain during injection. In this study, we compared the effectiveness of ondansetron, ramosetron, and granisetron, the three common 5-hydroxytryptamine 3 receptor antagonists in reducing the incidence and severity of propofol-induced pain (PIP).

Methods:

This study included 120 adult patients (18–65 years old) scheduled for elective laparoscopic cholecystectomy under general anesthesia. Patients were randomly assigned to four groups (n = 30 for each group). Granisetron (2 mg) was administered to Group G, ramosetron (0.3 mg) to Group R, ondansetron (4 mg) to Group O, and normal saline to Group N. Venous circulation was occluded by a rubber tourniquet to retain the drug in the vein and prevent its spread in circulation. Pain was assessed at 0, 5, 10, 15, and 20 s using a four-point verbal rating scale (VRS). The Kruskal–Wallis, Chi-square, Friedman tests, and generalized estimating equations were used to analyze the data.

Results:

Groups O, R, and G demonstrated significantly reduced pain compared to group N at all-time points (P < 0.001). Granisetron (Group G) was the most effective, with a mean VRS of 0.00 at 20 s. PIP incidence was 100% in Group N, compared to 80%, 70%, and 26.7% in Groups O, R, and G, respectively (χ² =40.443, P < 0.001).

Conclusion:

Granisetron is the most effective in reducing PIP, though ondansetron and ramosetron also offer significant pain relief.

INTRODUCTION

Propofol has a quick onset, short duration, and favorable side effect profile. However, pain during intravenous injection remains a significant issue, occurring in 28%–90% of patients.[1] Propofol-induced pain (PIP) is believed to result from direct vascular irritation or the release of inflammatory mediators like bradykinin, which sensitizes pain receptors.[2]

Various techniques have been suggested to decrease the discomfort caused by propofol injections with varying success. These include using different propofol formulations – such as emulsions and prodrugs – and pretreating with drugs such as ketamine, lignocaine, dexamethasone, opioids, and 5-hydroxytryptamine 3 receptor antagonists (5-HT3). 5 HT3 receptor antagonists are being utilized more often in anesthetic procedures to lower PIP.[1,3,4]

A meta-analysis published in 2016 verified their utility as pretreatments in lowering PIP.[5]

No studies have directly compared 5-HT3 antagonists for preventing propofol injection pain (PIP), though each is effective individually. Ondansetron was evaluated with normal saline,[6] ramosetron with lignocaine,[4] and granisetron with normal saline[7] for the prevention of PIP and was found to be efficacious. Laparoscopic cholecystectomy was chosen for this study to ensure homogeneity of the sample and because of the high usage of propofol in this procedure. As a common and standardized surgical procedure performed under general anesthesia, it minimizes variability in surgical and anesthetic factors, ensuring that observed differences in pain outcomes are attributable to the interventions rather than procedural inconsistencies. Hence, we conducted this study to compare these three drugs against each other and normal saline to determine the most effective option for reducing PIP incidence and severity.

METHODS

This study was designed to be a randomized, placebo-controlled trial. It was also double-blinded with parallel groups. The trial received approval from the Institutional Ethics Committee and was subsequently registered prospectively with the Clinical Trials Registry of India (CTRI) (CTRI/2021/11/038172). The study included 120 patients, aged 18–65 years, with American Society of Anesthesiologists (ASA) physical status I or II, of both genders, scheduled for elective laparoscopic cholecystectomy. Written informed consent was obtained from all participants before their inclusion. This study excluded patients who were allergic to propofol or 5HT3 receptor antagonists, had significant cardiac or conduction problems, difficulty in breathing and seizure disorders, or head injuries. Before surgery, all patients underwent a thorough preanesthetic evaluation, accompanied by the requisite diagnostic tests. The patients were kept nil by mouth for 6 h and were given orally 0.25 mg of alprazolam and 150 mg of ranitidine the night before and 2 h before induction.

Upon the patient’s arrival in the operating room, a multipara monitor was attached. An electrocardiogram, pulse oximeter, and noninvasive arterial blood pressure were used to get the baseline data. A prominent vein on the dorsum of the nondominant hand was used to insert one of the two 20-gauge intravenous cannulas. The cannula was connected to an infusion of 0.9% normal saline solution. We administered the study drugs through this cannula. The other intravenous cannula was utilized to deliver the maintenance fluid.

Patient allocation into four groups was performed using a computer-generated random number sequence to ensure unbiased distribution. In this study, 30 patients were randomly assigned to each of the following four groups: Group G (Granisetron 2 mg, Grandem, Aristo Pharmaceuticals Pvt. Ltd., India), Group O (Ondansetron 4 mg, Emigo, Zuventus Healthcare Ltd., Mumbai, India), Group R (Ramosetron 0.3 mg, Nozia, Cadila Healthcare Ltd., Goa, India), and Group N (0.9% normal saline, serving as the control group). The study drugs were diluted to a final injection volume of 4 mL with 0.9% normal saline by an anesthesiologist not involved in the study, ensuring blinding during the observation and data recording process. Venous circulation was occluded in the nondominant hand using a rubber tourniquet at the forearm to retain the drug inside the vein. The study drugs were given through the intravenous cannula over a 10-s interval. After 1 min, the occlusion was relaxed, and a 5 mL syringe was used to deliver 25% of the predicted induction dose (2 mg/kg) of propofol-LCT (long-chain triglycerides) in 20 s. A stopwatch was utilized to regulate the injection rate. The patient did not get any analgesics before the propofol infusion. The verbal rating scale (VRS), having four points of grading pain severity, developed by McCrirrick and Hunter,[8] was employed for rating pain at 0, 5, 10, 15, and 20 s following propofol administration. In the McCrirrick and Hunter VRS, pain is graded on a score from 0 to 3. Score 0 indicates no pain, while score 1 represents mild pain, where patients respond to questions about pain but do not show obvious signs like facial grimacing or crying. Score 2 refers to moderate pain, with clear indicators such as facial grimacing, hand withdrawal, or spontaneous complaints of pain. A score of 3 was given for patients showing strong voluntary vocal response along with motor activity like arm withdrawal, facial grimacing, or tears suggesting severe pain. The time at which propofol was administered was 0 s.

Using the VRS, an anesthesiologist who was unaware of the patients’ group assignments evaluated each patient’s degree of discomfort. After assessing discomfort, the remaining propofol-LCT dosage was delivered to induce the patient, and atracurium (0.5 mg/kg) was administered to facilitate endotracheal intubation. As an analgesic, fentanyl (2 mcg/kg) was administered. Vital signs were monitored, and isoflurane and intermittent atracurium were administered to maintain anesthesia. After the procedure, the patients were treated with neostigmine (50 µg/kg) and glycopyrrolate (12.5 µg/kg) and extubated. An anesthesiologist who was blinded examined the patient for signs of any allergic reaction, discomfort, or edema at the site of injection.

The sample size was estimated using the prior study’s assumptions, which included a 1.5-point difference in VRS scores, a standard deviation of 2, 80% power, and a significance threshold of 0.05. As a result, the computed sample size was 28 per group. We created a sample size of 30 patients per group to ensure adequate power and consider potential dropouts.

Statistical Analysis: The data were analyzed using IBM Corp.’s Statistical Package for the Social Sciences (SPSS) version 23 (Armonk, New York, USA). Based on the VRS, the mean pain scores for each of the four groups were assessed using analysis of variance. Due to potential non-normal distribution, the Kruskal–Wallis test was applied. The Chi-square test compared the percentage of patients with painless injections (VRS = 0) across groups. Changes in VRS within each group were analyzed using the Friedman test, while generalized estimating equations (GEE) were employed to compare changes in VRS over time between the different groups. P <0.05 indicated statistical significance for all comparisons.

RESULTS

All four groups were comparable in terms of patient characteristics and ASA physical status [Table 1], which represents that the four groups are similar inpatient characteristics and ASA status, supporting that any observed differences in outcomes are not due to baseline differences

Table 1.

Baseline characteristics of patients in the four groups

Patient characteristics	Group N (n=30)	Group O (n=30)	Group R (n=30)	Group G (n=30)	P
Age (years)€	40.67±13.28	43.07±13.21	37.50±11.38	38.40±12.08	0.317
Sex (male/female)¥	7/23	10/20	7/23	4/26	0.34
Weight (kg)€	63.13±9.32	66.13±8.15	64.03±8.64	63.80±8.58	0.442
ASA (I/II)¥	24/6	24/6	23/7	22/8	0.913

^€One-way ANOVA, ^¥Chi-square test. Values are expressed as mean±SD or number of patients. Group G=Granisetron (2 mg), Group R=Ramosetron (0.3 mg), Group N=Normal Saline (4 ml), Group O=Ondansetron (4 mg). n=number of patients, ASA=American Society of Anesthesiologists, SD=Standard deviation

Table 2 and Figure 1 present the VRS results at different time intervals, highlighting significant differences in pain relief across the groups at 5, 10, 15, and 20 s.

Table 2.

Verbal Rating Scale scores for pain relief at different time intervals across the four treatment groups

VRS€	Groups				P
	Group N (n=30), mean±SD	Group O (n=30), mean±SD	Group R (n=30), mean±SD	Group G (n=30), mean±SD
0 s	0.00±0.00	0.00±0.00	0.00±0.00	0.00±(0.00)	-
5 s	1.23±0.77	0.27±0.69	0.10±0.31	0.00±0.00	<0.001
10 s	1.63±0.67	0.63±0.81	0.13±0.35	0.00±0.00	<0.001
15 s	1.97±0.72	1.03±0.81	0.33±0.48	0.03±0.18	<0.001
20 s	2.23±0.73	1.13±0.82	0.73±0.52	0.27±0.45	<0.001
P-value for change in VRS over time within each groupβ	<0.001	<0.001	<0.001	<0.001
Overall P-value for comparison of the change in VRS over time between the four groups	<0.001

^€Kruskal–Wallis test, ^βFriedman test; Values are expressed as mean±SD. Group G=Granisetron (2 mg), Group R=Ramosetron (0.3 mg), Group N=Normal Saline (4 ml), Group O=Ondansetron (4 mg). GEE=Generalized estimating equations, SD=Standard deviation, VRS=Verbal Rating Scale

Figure 1.

Time-dependent variations in Verbal Rating Scale (VRS) Scores. Group G: Granisetron (2 mg), Group R: Ramosetron (0.3 mg), Group N: Normal Saline (4 ml), Group O: Ondansetron (4 mg)

• At 5 s, group N had a mean VRS of 1.23 ± 0.77, while Group O had 0.27 ± 0.69, Group R had 0.10 ± 0.31, and Group G had 0.00 ± 0.00. Groups O, R, and G experienced significantly better pain relief than Group N (P < 0.001)

• At 10 s, the mean VRS was 1.63 ± 0.67 for Group N, 0.63 ± 0.81 for Group O, 0.13 ± 0.35 for Group R, and 0.00 ± 0.00 for Group G. Groups O, R, and G again showed superior pain relief compared to Group N (P < 0.001)

• At 15 s, group N had a mean VRS of 1.97 ± 0.72, Group O had 1.03 ± 0.81, group R had 0.33 ± 0.48, and Group G had 0.03 ± 0.18. A statistically significant difference was observed when comparing group N with the other groups (P < 0.001)

• At 20 s, Group N had a mean VRS of 2.23 ± 0.73, Group O had 1.13 ± 0.82, Group R had 0.73 ± 0.52, and Group G had 0.27 ± 0.45. Significant differences in pain relief were also noted in the other groups compared to Group N (P < 0.001).

These results demonstrate that groups O, R, and G experienced significantly better pain relief at each time point compared to Group N, with Group G showing the greatest reduction in pain. Each group showed a significant reduction in pain over time, as indicated by the Friedman test (P < 0.001 for all groups), demonstrating the effectiveness of each treatment. The GEE method revealed a significant overall difference in VRS changes between the groups (P < 0.001), indicating varying degrees of pain relief among the treatments [Table 2].

Table 3 illustrates how the incidence of pain differed considerably between the groups. In group N, all participants (100%) reported pain, compared to 80% in group O, 70% in group R, and 26.7% in group G. Chi-square test confirmed a statistically significant difference in pain scales amongst the groups (χ² =40.443, P < 0.001). Comparative analysis with adjusted P values showed that group G had a significantly lower pain incidence compared to groups O, R, and N (P < 0.001 for all comparisons). The incidence of discomfort was also considerably lower in groups O and R than in group N (P = 0.028 and 0.003, respectively). Nevertheless, there was no significant difference in the occurrence of pain between groups O and R (P = 0.552). Overall, the results highlight the varying effectiveness of the treatments in reducing pain, with Group G showing the best outcome in minimising the incidence of pain.

Table 3.

Pain incidence analysis among different treatment groups

Any pain	Group					Chi-squared test
	N, n (%)	O, n (%)	R, n (%)	G, n (%)	Total, n (%)	χ 2	P
Yes	30 (100.0)	24 (80.0)	21 (70.0)	8 (26.7)	83 (69.2)	40.443	<0.001
No	0	6 (20.0)	9 (30.0)	22 (73.3)	37 (30.8)
Total	30 (100.0)	30 (100.0)	30 (100.0)	30 (100.0)	120 (100.0)
Group			Adjusted P
G versus O			<0.001
G versus R			0.003
G versus N			<0.001
O versus R			0.552
O versus N			0.028
R versus N			0.003

Chi-square test, Group G=Granisetron (2 mg), Group R=Ramosetron (0.3 mg), Group N=Normal Saline (4 ml), Group O=Ondansetron (4 mg)

DISCUSSION

This study found that all three study drugs were able to reduce PIP compared to normal saline, with granisetron providing the most substantial pain relief. These findings align with previous research that highlighted the analgesic potential of 5-HT3 receptor antagonists for the prevention of PIP.[4,5,9,10] While various studies have evaluated 5-HT3 antagonists like ondansetron, ramosetron, and granisetron against other medications for preventing PIP, there has been no direct comparison of these drugs with one another for this specific purpose.

Various methods have been explored to mitigate PIP, with differing levels of effectiveness. These methods include diluting the propofol or increasing the infusion rate. In addition, pharmacologically active agents have been co-administered and studied. They include lignocaine, aspirin, opioids, ephedrine, ondansetron, or metoclopramide.[3,11,12,13]

Ondansetron, ramosetron, granisetron, dolasetron, palonosetron, alosetron, and tropisetron are known as 5-HT3 antagonists. Numerous studies have examined how well a number of these medicines work to lower PIP.[14]

It is believed that the main mechanism underlying PIP is the release of bradykinin by propofol via the stimulation of the Kallikrein–Kinin pathway. This process causes veins to be dilated along with increased vascular permeability. This increases the interaction and contact among nerve endings with the aqueous phase of propofol. Consequently, this interaction contributes to the sensation of pain associated with propofol injection.[15] 5-HT3 receptor antagonists may interact with μ-opioid receptors, exhibiting agonistic effects that contribute to their analgesic properties.[16] Similarly, Palonosetron is unique from other 5-HT3 receptor antagonists by its unique chemical structure, greater binding affinity (pKi = 10.45), and substantially longer half-life (40 h).[17] In addition, since 5-HT3 receptors are implicated in nociceptive pathways, their blockade could further enhance the analgesic effects of these drugs.[16]

In our placebo-controlled trial, we focused on a comparison of ondansetron, ramosetron, and granisetron with one another in minimizing PIP. When compared to normal saline, the results indicate a considerable decrease in pain after propofol administration after pretreatment with two widely used antiemetic medications. A previous study revealed that combining granisetron with venous occlusion for 1 min successfully reduced PIP. This finding highlights granisetron’s potential in pain management.[7] However, it is noteworthy that while the previous study demonstrated the effectiveness of granisetron, it did not compare the analgesic effects of granisetron with ondansetron and ramosetron. Our study seeks to fill this gap by systematically comparing these agents to determine if significant differences exist in their ability to reduce PIP.

In contrast to the control group, our study shows that ondansetron (4 mg) significantly reduces propofol injection pain, aligning with the results of Wasinwong et al.[18] In their study, the ondansetron group experienced discomfort 82.5% of the time, compared to 85% in the control group. In our trial, 80% of patients in the ondansetron group reported pain, compared to 100% in the control group, indicating a higher incidence of pain but highlighting the preventive benefit of ondansetron. Similarly, Singh et al.[4] observed that ramosetron (0.3 mg) decreased the incidence of PIP from 65% in the control group to 30%. Ramosetron’s ability to reduce pain related to propofol injection was further supported by the fact that our study showed a 30% decrease in pain incidence when compared to the control.

Lee et al.[19] reported 60% pain in the ramosetron pretreatment group. In comparison, our study observed a slightly higher incidence of 70%, suggesting some variability in response to ramosetron across different patient populations. Both findings suggest that ondansetron and ramosetron significantly reduce PIP, but with variations in the extent of pain prevention across studies.

According to a comparison study by Hwang et al.,[20] 0.3 mg of ramosetron was superior to 4 mg of ondansetron in terms of managing PIP. However, in contrast to their findings, our study showed no significant statistical difference between ramosetron and ondansetron. This discrepancy could be attributed to differences in methodology, as they manually occluded the forearm after administering the pretreatment drug, while we used a tourniquet.

In the study by Ahmed et al.,[7] granisetron was found to reduce PIP from 60% to 15%. However, these findings are inconsistent with our current study, where pretreatment with granisetron reduced the incidence of pain from 100% to 26.7%.

This discrepancy may be attributed to several factors: granisetron has a higher propensity to attach to 5-HT3 receptors compared to ondansetron and ramosetron, along with a longer half-life and a stronger vasodilatory effect. In addition, granisetron is metabolised differently from ondansetron and ramosetron, which could result in less variability in its effects among patients. This higher affinity not only enhances its analgesic effects but also makes granisetron less susceptible to interindividual variability, as it maintains consistent efficacy across a range of patient populations. As such, granisetron may exert a more significant impact not only peripherally but also centrally.[21,22]

Our comparison revealed that granisetron was the most effective 5-HT3 antagonist, reducing the occurrence of PIP from 100% to 26.7%, while ramosetron and ondansetron lowered the incidence to 70% and 80%, respectively.

The study has several limitations that may affect the generalizability of the findings. First, a larger cohort could provide more robust data. The focus on a specific procedure may limit the generalizability of these findings to other patient populations receiving propofol. In addition, using a VRS introduces potential bias based on individual pain tolerance. The evaluation was limited to immediate postinjection responses, without considering longer-term pain, and the method of drug administration could inadvertently influence reporting due to unintentional bias. Variability in individual responses and the exclusion of certain populations also limit applicability.

CONCLUSION

This study demonstrates that granisetron is the most effective 5-HT3 antagonist in reducing the incidence and severity of PIP, followed by ramosetron and ondansetron, all of which showed significant efficacy compared to normal saline. These findings highlight the dual benefits of 5-HT3 antagonists as both antiemetics and analgesics. Granisetron’s superior efficacy may be attributed to its higher receptor affinity and longer half-life, offering potential advantages in clinical practice. However, further multicenter trials involving diverse surgical procedures and patient populations are needed to validate and expand upon these results.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.