The Effect of Postoperative Analgesia on the Day-Case Rate of Laparoscopic Cholecystectomy: A Randomised Pilot Study of the Laparoscopic-Assisted Right Subcostal Transversus Abdominis Plane Block plus Local Anaesthetic Wound Infiltration versus Local Anaesthetic Wound Infiltration only

Laparoscopic cholecystectomy (LC) is nowadays performed on a day-case (DC) basis in many centres worldwide. Day surgery has well-established advantages for the patients and the healthcare system alike; however, the postoperative pain remains one of the main factors affecting patients’ discharge. The combination of infiltration of local anaesthetic infiltration (LAI) into the port sites and the prescription of oral analgesia has been regarded as the standard of care for the management of postoperative pain after LC [1]. The transversus abdominis plane (TAP) block has been introduced as a new technique to improve pain control. This was first described in 2001 [2] and consisted of a peripheral nerve block designed to anaesthetise the nerves supplying the anterior abdominal wall (T6 to L1). The procedure is performed either under ultrasound guidance or direct laparoscopic visualisation, and the current evidence supports its beneficial role in LC [3]. The TAP block and LAI seem to be equally effective in postoperative pain control following LC [4]; however, most of the comparative studies assessed patients who received either the TAP block or LAI, and the regional block was often performed with larger volumes of local anaesthetic and multi-site injections of the abdominal wall [5]. Little data exist on the association of single-injection TAP block + LAI versus LAI alone, and the impact of the two techniques on the DC rate of LC has not been investigated. While this gap would make a randomised controlled trial (RCT) desirable, the uncertainty about effective pain control and participants’ acceptance represents a potential area of concern. For this reason, in this pilot study we aimed to assess whether the research design was appropriate to conduct a larger RCT on the role of the laparoscopic right subcostal TAP block on postoperative pain, after LC.

The research was designed as a prospective, single-blinded, parallel randomised controlled study, that was conducted according to the Declaration of Helsinki and Good Clinical Practice guidelines. The National Research Ethics Service (NRES) Committee, Southwest-Bristol, UK, approved the study (18/SW/0147), and it was registered in the ClinicalTrials.gov registry (NCT03532906, May 17, 2018). All the trial participants provided written informed consent.

The power analysis was conducted under the assumption of a large effect between treatment groups. A sample of 60 participants, 30 per arm, was estimated for two-sided α = 0.05%, power of 80%, to detect a 75% difference in early postoperative pain scores. Since pain is a subjective symptom, the comparison between individuals would be difficult to quantify. If small differences in pain scores were observed, they might not necessarily be related to the anaesthetic technique used but rather be associated with other factors (i.e., the individual response to pain medications). Instead, if a large difference was detected, that might be interpreted as a relevant clinical effect. Patients aged 18–60, who were scheduled for elective DC LC, were considered for the study. Exclusion criteria were emergency LC, planned postoperative in-patient stay, conversion to open cholecystectomy, contraindications to administration of bupivacaine and/or any of the standard oral painkillers prescribed postoperatively, history of chronic pain and/or consumption of painkillers prior to surgery, unwillingness to give informed consent for participation to the study.

Subjects who were eligible were randomised in two groups: subcostal TAP block plus local anaesthetic wound infiltration (STALA) and LAI only. Simple randomisation was achieved with 1:1 allocation ratio and was generated with Microsoft^® Excel^® 2016 (Microsoft, Redmond, WA, USA); each subject was allotted a sequentially numbered, opaque, sealed envelope containing the randomisation assignment arm, which was opened during surgery. The study participants and the statistician who generated the randomisation sequence were blinded to group allocation.

All the subjects underwent the same general anaesthetic protocol – intravenous (IV) propofol 1.5 mg/Kg, IV fentanyl 2 μg/Kg, atracurium 0.5 mg/Kg, for induction, nitrous oxide and sevoflurane for maintenance. They all received intraoperative anti-emetic prophylaxis, which consisted of IV ondansetron 4 mg and IV dexamethasone 3.3 mg. Two syringes were prepared before the start of the procedure: one contained bupivacaine 0.5% 20 mL, and one bupivacaine 0.5% 10 mL. LC was performed with the standard 4-trocar technique via two 12 mm incisions at the umbilicus and epigastrium, two 5 mm incisions in the right upper quadrant and right flank. The right subcostal TAP block consisted of a single injection performed at the end of surgery, under direct visualisation of the right transversus abdominis muscle; a 40-mm, 21-gauge BD Eclipse™ needle (BD, Franklin Lakes, New Jersey, USA) was advanced percutaneously, perpendicular to the abdominal wall, at the level of the right anterior axillary line. The correct plane of infiltration was confirmed by the formation of the Doyle’s bulge [6] as shown in Figure 1. Both cohorts received 30 mL bupivacaine: the STALA group was given 20 mL for the TAP block injection, 10 mL for the infiltration of the midline port sites (umbilical and epigastric). The LAI group received 20 mL for the midline port sites and 10 mL for the right lateral ones (right upper quadrant and right flank). After extubation, subjects were transferred to the postoperative recovery unit (PORU) and received the same standard analgesic protocol consisting of oral paracetamol 1 g every 6 h and oral ibuprofen 400 mg every 8 h, while opioids were given as rescue analgesia [6]; namely, patients received oral codeine 60 mg every 6 h, if the pain score on the Visual Analogue Scale (VAS) [7] was 4–5, or IV fentanyl 20 µg every 5 min, if VAS was >5. Subjects whose pain was not relieved by 2 consecutive fentanyl doses were given IV or oral morphine 5–10 mg every 2 h. Although patients were not formally scored for postoperative nausea and vomiting in the preoperative setting, those with history of postoperative nausea and vomiting were given prophylactic IV/oral cyclizine 50 mg and IV/oral ondansetron 4 mg in the PORU. Subjects whose vital signs remained stable and whose symptoms were adequately controlled were deemed fit for discharge; they were given the same standard prescription of oral painkillers received in the PORU, for a duration of 7 days. Participants’ well-being was assessed on postoperative day 3 and 7, by means of a telephone interview based on the Quality of Recovery-15 (QoR-15) questionnaire [8]. At the time, the oral intake of painkillers was evaluated as well. The interviewer was blinded to patients’ group allocation. Data from subjects who answered one or both questionnaires were included in the study. After the second telephone interview, no further follow-up was conducted.

VAS scores were assessed at 1 h after surgery and at discharge. The postoperative pain scores were also categorised in mild (VAS <4), moderate (VAS 4–6), strong (VAS 7–8), severe (VAS 9–10) [9]. The length of stay in the PORU was defined when subjects whose vital signs remained stable were deemed as clinically fit to be transferred to the DC ward.

Primary objective was the acceptability of the postoperative pain scores as measure of efficacy of the interventions in a definitive trial; the adequacy of the VAS scores was assessed towards the need for IV opioids in the PORU and the DC rate (i.e., the number of subjects who were discharged on the same day of surgery over the total number of patients who underwent LC). Secondary objective was the participants’ adherence to follow-up, which was evaluated as the number of subjects who answered one or both phone calls over the total number of patients who underwent LC.

Statistical analysis was conducted using Prism^®, version 9.3.1 (GraphPad Software, San Diego, CA, USA). Continuous variables with a normal distribution were described using the mean ± standard deviation; means between groups were compared with the Student’s t test. Data that were not normally distributed were described as median; medians were compared with the Mann-Whitney U test. Categorical data were given as absolute numbers and percentages, and difference in frequencies was evaluated with the Fisher’s exact test. Two-tailed p values were used and were considered as significant if <0.05.

The trial occurred between 2018 and 2021 and was put on hold in 2019, due to a reconfiguration of departmental elective surgical schedules, and in 2020, due to the SARS-CoV-2 (COVID-19) pandemic. Overall, 110 subjects were approached and 60 were included in the study: 29 were randomised to STALA, and 31 to LAI as shown in Figure 2.

Symptomatic gallstones were the indication of surgery in all the study participants. Patients’ baseline characteristics, total duration of surgery and anaesthesia, intraoperative technical difficulty based on the Nassar’s classification [10], and postoperative outcomes are described in Table 1. There were 51 (85%) women, and mean age was 41.3 ± 11.3 years. Subjects in the LAI group were all women (p = 0.0007) and younger than the STALA – 37.7 ± 10.25 versus 43.8 ± 10.50 years (p = 0.023). No postoperative complications occurred. The median VAS scores after STALA and LAI were 0 (range 0–5) versus 1 (range 0–6) at 1 h (p = 0.60) and 0 (range 0–3) versus 1.5 (range 0–2), at discharge (p = 0.55; Fig. 3). Overall, 28 (46.7%) participants received opioids before discharge – 15/29 (51.7%) in STALA versus 13/31 (41.9%) in LAI (p = 0.60). The overall DC rate was 93.1% (27/29) after STALA and 93.5% (29/31) after LAI (p = 0.39). Six patients (4 in the STALA group and 2 in LAI) remained in hospital because of postoperative pain (3), nausea (1), challenging surgical procedure (1), and late theatre schedule (1), respectively. Of those with poorly controlled pain, 1 received the STALA, and 2 had LAI. One subject in the STALA group was readmitted because of persistent nausea, 2 days post-surgery. All the 7 patients who required in-patient stay or early readmission were excluded from the postoperative follow-up. Telephone interviews were attempted in 53 subjects, and 3 provided no answer; 50 (83.3%) participants responded to one or both questionnaires – 49 (92.5%) on postoperative day 3 and 45 (84.9%) on day 7. The QoR-15 answers are summarised in Table 2 and Table 3. At day 3, subjects in the STALA group provided higher median scores in the ability to look after themselves (10 vs. 8; p = 0.03). Of the 46 (93.9%) who had regular analgesia, 19 (42.2%) took codeine, with 10 in the STALA and 9 in LAI group (p = 0.38). At day 7, 13 (28.9%) subjects reported intake of oral analgesia, with 3 (23.1%) having opioids (1 in STALA and 2 in the LAI group).

No unintended consequences occurred with the interventions.

Effective postoperative pain control is key for the provision of efficient day surgery services. The major goal of pain management should be the prescription of adequate analgesia with the minimum dosage of medications to reduce side effects; however, the optimal regimen is still debated [11]. Although standardised protocols are used in common practice, effective postoperative pain management should be tailored to the needs of the individual patients. LC is associated with moderate to severe postoperative pain that is largely caused by the wound incisions [12]. For this reason, the infiltration of local anaesthetic via LAI or TAP block plays a key role in pain relief [13]. In the TAP block, the plane of infiltration between the internal oblique and transverse muscles is minimally vascularised; therefore, the analgesic effect can last up to 24 h [14], while that of LAI lasts 6–8 h [15]. In the published literature, the comparison of TAP block versus LAI has led to conflicting results. Two systematic reviews reported marginally better postoperative pain scores after TAP block, but the authors acknowledged that results were limited by the variation of techniques used for the loco-regional block, the type, dose, and volume of local anaesthetic used [16, 17]. Conversely, other authors reported better outcomes following LAI [18], while another study reported similar effects between TAP block and LAI [19]. The effectiveness of the TAP block relies on the volume rather than the concentration of the local anaesthetic [20], particularly if the injection is conducted in multiple areas of the abdominal wall [21]. Larger volumes administered via multi-site injections would probably allow for a better distribution of local anaesthetic to the abdominal plane of innervation. We conducted a pilot study comparing the association of the right subcostal TAP block performed under direct laparoscopic visualisation and LAI versus LAI only and assessed the appropriateness of the research design in view of a larger RCT. Bupivacaine was chosen as per our institutional standard protocol. The TAP block was performed at the end of LC, following some evidence that it was more effective than when performed before surgery [22]. Given that the right subcostal injection alone seems to provide an incomplete block of T10 [23] and the TAP block has a failure rate of 5–20% [24], we were concerned that participants would suffer from poorly controlled postoperative pain in the midline wound incisions. For this reason, we added the infiltration of a small volume of local anaesthetic to the midline port sites. Both groups received the same total volume and concentration of local anaesthetic. Patients in the LAI group were females and younger than those in the STALA; we believe that the random allocation of male patients in one cohort only was due to chance. Our preliminary results showed that both the postoperative pain scores at 1 h and at discharge and the need of IV opioids in the PORU did not differ between the two study groups. In particular, rescue analgesia with IV fentanyl was needed in 10 subjects in the STALA group and 11 in LAI. Moreover, patients who received STALA remained longer in PORU, though the difference was not statistically significant. Similar results were reported by Siriwardana et al. [18], who authored the only published study on the laparoscopic right subcostal TAP + LAI versus LAI only. The authors found no differences in postoperative pain control after LC; however, they excluded patients who underwent technically complex LC and did not describe the surgical DC rate. In our series, we included both straightforward and technically complex LC, to evaluate the postoperative pain control in real-life circumstances. Also, we assessed the two anaesthetic techniques towards the DC rate, as an indirect sign of adequate pain control. To the best of our knowledge, the role of the TAP block and LAI on the DC rate of LC has not been assessed before. Ninety-three percent of the participants were discharged on the same day of surgery; nevertheless, such a result was expected, given that the study included patients who were scheduled for DC LC only. Nonetheless, 3 (5%) subjects were not discharged because of uncontrolled pain, and interestingly, all of them had mild pain score; while it could be argued that a discrepancy about subjective and objective description of symptoms exists, such a low rate could potentially be interpreted as a measure of adequate postoperative analgesia.

Our results suggest that the effective analgesia achieved with the combination of LAI and postoperative medications allows for a high DC rate of LC, while the right subcostal TAP block does not seem to add benefits to it. However, we cannot infer definitive conclusions on the subject, because of the small sample size and the evidence that same-day discharge rates can be affected by factors other than pain (i.e., postoperative complications, side effect of medications, social reasons). In most of the comparative studies that reported the superiority of the TAP block over LAI, the regional block was performed under ultrasound guidance; given that the local anaesthetic is delivered correctly only in 23% of cases [25], it has been suggested that the TAP block under ultrasound guidance could be more effective than the blind injection and the laparoscopically guided technique [26]. This aspect could in part explain why in our series, the TAP block did not seem to add benefits to LAI. Nevertheless, another study on the TAP block reported similar effectiveness under both ultrasound and laparoscopic guidance [27]. Another possible reason why in our series the association of TAP block + LAI was as effective as LAI alone resides on the effectiveness of the postoperative analgesic regimen. In fact, it has been reported that in the presence of adequate multimodal analgesia, the TAP block yields minimal to no advantage in the postoperative pain management [28]. The participation to the telephone interviews was high, showing patients’ adherence to the follow-up, and the QoR-15 answers described good postoperative recovery in both groups. It has been reported that neither the single nor the multi-sided TAP block injection impacts on the postoperative quality of recovery following LC [29]; this aspect seems to corroborate the importance of the multimodal analgesia on patients’ post-surgical well-being.

Potential criticisms to the study are the single-blinded design, the different demographic characteristics of the two groups, the inclusion of participants with no contraindication to any of the postoperative pain medications, and the different volumes of local anaesthetic used in the midline port sites. We conducted a single-blinded trial for practical reasons; in order to perform double blinding, we could have involved two different surgeons performing LC and the TAP block independently or alternatively, and we could have used the TAP block with injection of a placebo (i.e., 0.9% sodium chloride) in the control group. Having two different surgeons would be impractical, whereas using TAP block with placebo could potentially be associated with pain secondary to the infiltration of the muscle-aponeurotic plane, thus acting as a confounding factor [30]. We chose to include working age participants only, hypothesising that the short-stay surgery might have a greater relevance to their activity levels and time required off work, than older subjects. Nonetheless, age cut-off and different sex characteristics between groups would prevent the generalisability of results. The inclusion of participants with no contraindications to pain relief medications allowed for a homogeneous postoperative treatment protocol; in real conditions, patients’ tolerance to each of the pain medications varies greatly and this could be associated with different pain scores. However, setting up a study with various pain control regimens would be impractical. Finally, the injection of different volumes of local anaesthetic in the midline wound incisions could potentially bias results. If all the participants received 30 mL of local anaesthetic with the same volume in the midline trocars, the LAI group would receive only 10 mL at the umbilical and epigastric wounds and such a small dose could potentially result in suboptimal pain control.

In this pilot study, the laparoscopically guided right subcostal TAP block did not provide additional benefit to LAI and our results suggested that LAI and the postoperative pain medications may play a major role in the postoperative analgesia and high DC rate of LC. For these reasons, despite the appropriate design, our findings do not support a larger RCT.

The authors would express their gratitude to Mr Edoardo Ricciardi, Mrs Hazel Bennett, and Miss Emma Palmer, for their invaluable support with the administrative work.

The research was conducted according to the Declaration of Helsinki and Good Clinical Practice guidelines. The National Research Ethics Service (NRES) Committee, Southwest-Bristol, UK, approved the study (18/SW/0147), and it was registered in the ClinicalTrials.gov registry (NCT03532906, May 17, 2018). The research protocol was reviewed and approved by the Health Research Authority (HRA) and Health and Care Research Wales, study number 245942. All the trial participants provided written informed consent.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Davide Di Mauro: study conception and design, data acquisition and interpretation, drafting the work, approval of the final version of the manuscript, and agreement to be accountable for all aspects of the work. Alex Reece-Smith: data acquisition and interpretation, critical revision of the work for important intellectual content, approval of the final version of the manuscript, and agreement to be accountable for all aspects of the work. Ikechukwu Njere: data analysis and interpretation, critical revision of the work for important intellectual content, approval of the final version of the manuscript, and agreement to be accountable for all aspects of the work. Sheena Hubble: data interpretation, critical revision of the work for important intellectual content, approval of the final version of the manuscript, and agreement to be accountable for all aspects of the work. Antonio Manzelli: study design, data acquisition and interpretation, critical revision of the work for important intellectual content, approval of the final version of the manuscript, and agreement to be accountable for all aspects of the work.

The study protocol and datasets generated and analysed during the current trial are not publicly available, in accordance with the Confidentiality Policy in force at the Royal Devon University Healthcare NHS Foundation Trust; this is a legal requirement under the Data Protection Act 2018 (UK), but data are available from the corresponding author, on reasonable request.