Programmed Intermittent Bolus versus Continuous Infusion for Regional Analgesia Following Thoracoscopic Surgery: A Systematic Review and Meta-Analysis

Introduction

Thoracic surgery, including lung resection, esophagectomy, and mediastinal tumor resection, is often associated with severe postoperative pain due to the extensive dissection of intercostal nerves and pleural surfaces.¹ Thoracoscopic surgery has become a mainstay for various thoracic conditions, offering advantages such as reduced pain, shorter hospital stays, and improved cosmetic outcomes compared to traditional open thoracotomy.² However, postoperative pain management remains a significant challenge, impacting patient recovery and satisfaction.¹ Effective pain management is essential not only for patient comfort but also for promoting early mobilization, reducing pulmonary complications, and improving overall recovery.³

Conventional methods, such as intravenous opioids and patient-controlled analgesia (PCA), are frequently used but are limited by their systemic side effects, including respiratory depression, nausea, vomiting, and pruritus.⁴ In recent years, regional analgesia techniques have gained popularity as adjuncts or alternatives to systemic analgesics.⁵ Among these, thoracic paravertebral block (TPVB), erector spinae plane block (ESPB), and serratus anterior plane block (SAPB) have emerged as valuable adjuncts to multimodal analgesia strategies for thoracic surgery.^6–8 By selectively blocking specific nerves, regional analgesia provide targeted pain relief while minimizing systemic opioid-related side effects such as nausea, vomiting, and respiratory depression.⁵

Single-injection nerve blocks have been demonstrated to provide effective analgesia for patients undergoing thoracoscopic surgery, but the duration of analgesia typically does not extend through the entire acute postoperative period. Consequently, clinical practices and trials involving continuous regional analgesia (CRA) are focused on improving pain control during the 12 to 24 hours and beyond. CRA involves the placement of a catheter near specific nerves and the administration of local anesthetics (LAs) over an extended period.⁹ This approach offers the advantage of prolonged analgesia, allowing for sustained pain control throughout the postoperative period.¹⁰

Among the various CRA techniques, two primary infusion strategies exist: continuous infusion (CI) and programmed intermittent bolus infusion (PIBI). CI delivers a constant flow of LAs, maintaining a stable concentration at the nerve site.¹¹ PIBI, on the other hand, involves periodic bolus injections of LAs, followed by a period of no infusion.¹¹ While both CI and PIBI have demonstrated efficacy in postoperative pain management, their relative advantages and disadvantages remain a subject of debate. PIBI may offer improved pain control during breakthrough pain episodes due to the bolus effect.¹² Conversely, CI provides more stable analgesia and it may be simpler to administer and require less frequent adjustments compared to PIBI.¹²

Despite their theoretical advantages, the evidence regarding their comparative effectiveness specifically in the context of thoracoscopic surgery remains limited and inconclusive. Given the increasing use of thoracoscopic surgery and the ongoing debate regarding optimal CRA strategies, a comprehensive synthesis of the available evidence is crucial to inform clinical practice. This meta-analysis aims to systematically review and quantitatively synthesize the results of randomized controlled trials (RCTs) comparing PIBI with CI for peripheral nerve block following thoracoscopic surgery.

Methods

Our study complies with the PRISMA statement and the protocol was registered in the International Prospective Register of Systematic Reviews (CRD420251002276).

Inclusion and Exclusion Criteria

We included studies according to PICOS criteria: patients undergoing thoracoscopic surgery (P); the infusion mode of peripheral nerve block is PIBI (I); the infusion mode of peripheral nerve block is CI (C); postoperative pain scores and opiates consumption within 48 hours, total LAs consumption, blocked dermatomes, and number of people using additional analgesics (O); RCTs (S). These researches would be excluded: (1) the data cannot be used for statistical analysis; (2) the data was incomplete or duplicate; (3) unpublished studies, crossover randomized design studies.

Search Strategy

We searched PubMed, Embase, Web of Science, and the Cochrane Library from inception to March, 2025. The search strategy was based on the PICOS framework and the key words, including “intermittent”, “continuous”, and “thoracoscopic”. There was no limitation on sample size, surgical types and language. The full search strategy of PubMed as representative was shown in supplementary material 1.

Data Extraction

All titles, abstracts, and full texts were reviewed by two reviewers individually. Any disagreements were resolved by the third reviewer. The data extracted from each RCT included baseline demographic information, information regarding assessment of the risk of bias, and outcomes. The primary outcome was resting pain score at 6 h. The secondary outcomes included postoperative resting and movement pain score, morphine consumption, LAs consumption, number of people using additional analgesics within 48 h, and range of sensory blockade (the number of thoracic cutaneous segments).

Certainty of Evidence

Two independent reviewers assessed the risk of bias by using the Cochrane Collaboration’s tool, which included of sequence generation, allocation sequence concealment, blinding of participants and personal, blinding of outcome assessment, incomplete outcome data, selective outcome reporting and other potential sources bias.¹³ The GRADE methodology was used to evaluate the quality of evidence for each outcome.

Statistical Analysis

Statistical analysis was carried out in ReMan 5.4 software. The data expressed as median and inter-quartile range was converted to mean and standard deviations,¹⁴ and opiates consumption was converted to intravenous morphine equivalent (mg).¹⁵ For the number of people using additional analgesics within 48 h, the pooled relative risk (RR) and 95% confidential intervals (CIs) was calculated. For other outcomes, the mean differences (MD) and 95% CIs were evaluated. All analyses were initially carried out using a fixed-effects model. Heterogeneity was regarded as significant if I² ≥ 50%, and the analyses were adopted with a random-effects model.¹⁶ Meanwhile, we performed sensitivity analysis to find the sources of heterogeneity. The funnel plot was used to exclude potential publication bias. The subgroup analysis was based on the types of nerves block. P value < 0.05 was considered as statistical significance throughout the study.

Results

The process of literature selection was shown in Figure 1. Seven studies enrolling 394 patients were included in the meta-analysis, between PIBI (n = 197 patients) and CI (n = 197 patients) groups.^17–23 A summary of the characteristics, interventions, and outcomes of each study included in our analysis was shown in Table 1. The most common types of nerve block included thoracic paravertebral block (TPVB) and erector spinae plane block (ESPB).

Table 1 Study Characteristics of Included Studies

Figure 1 PRISMA flow diagram of study selection. PRISMA figure adapted from Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Syst Rev. 2021;10(1):89.24

Primary Outcome

The resting VAS at 6 h was reported in 4 RCTs. The risk of bias was shown in Figure 2. Figure 3 suggested a low publication bias. The pooled data did not suggest that PIBI was superior to CI (MD = −0.52, 95% CI: −1.23 to 0.20; I² = 79%; Figure 4). The sensitivity analysis showed a similar result after deleting one RCT²³ (MD = −0.00, 95% CI: −0.00 to 0.00; I² = 49%; Figure 5). The subgroup analysis did not suggest that PIBI was superior to CI in TPVB group (MD = −0.44, 95% CI: −1.35 to 0.47; I² = 81%; Figure 6). A summary of the evidence was presented in Table 2.

Table 2 Secondary Outcomes for PIBI Versus CI

Figure 2 Risk of bias of resting VAS at 6 h (0–10 scale).

Figure 3 Funnel plot of resting VAS at 6 h (0–10 scale).

Figure 4 Forest plot of resting VAS at 6 h (0–10 scale). Abbreviations: PIBI, programmed intermittent bolus infusion; CI, continuous infusion.

Figure 5 Sensitivity analysis of resting VAS at 6 h (0–10 scale). Abbreviations: PIBI, programmed intermittent bolus infusion; CI, continuous infusion.

Figure 6 Subgroup analysis of resting VAS at 6 h (0–10 scale). Abbreviations: PIBI, programmed intermittent bolus infusion; CI, continuous infusion.

Secondary Outcomes

PIBI was found to be statistically more effective than CI in resting VAS at 12 h (MD = −0.71, 95% CI: −1.39 to −0.02; I² = 67%; Figure 7A), movement VAS at 12 h (MD = −0.48, 95% CI: −0.74 to −0.22; I² = 36%; Figure 7B), and morphine consumption in first 24 h (MD = −4.57, 95% CI: −7.78 to −1.35; I² = 11%; Figure 7C).

Figure 7 Continued.

Figure 7 (A). Forest plot of resting VAS at 12 h (0–10 scale). (B) Forest plot of movement VAS at 12 h (0–10 scale). (C) Forest plot of total morphine consumption in first 24 h (intravenous morphine equivalent, mg). (D) Forest plot of total postoperative local anesthetic consumption. (E) Forest plot of postoperative blocked dermatomes. (F) Forest plot of resting VAS at 24 h (0–10 scale). (G). Forest plot of resting VAS at 48 h (0–10 scale). (H) Forest plot of movement VAS at 6 h (0–10 scale). (I). Forest plot of movement VAS at 24 h (0–10 scale). (J) Forest plot of movement VAS at 48 h (0–10 scale). (K) Forest plot of total morphine consumption in second 24h (intravenous morphine equivalent, mg). (L) Forest plot of number of people using additional analgesics. Abbreviations: PIBI, programmed intermittent bolus infusion; CI, continuous infusion.

PIBI was found to have significantly lower LAs consumption (MD = −17.57, 95% CI: −34.13 to −1.01; I² = 66%; Figure 7D) and wider range of sensory blockade than CI (MD = 1.09, 95% CI: 0.87 to −11.31; I² = 0%; Figure 7E).

However, there were no significant difference between PIBI and CI in terms of the resting VAS at 24 h (MD = −0.20, 95% CI: −0.49 to 0.09; I² = 0%; Figure 7F), resting VAS at 48 h (MD = −0.41, 95% CI: −0.77 to −0.06; I² = 55; Figure 7G), movement VAS at 6 h (MD = −0.41, 95% CI: −1.00 to 0.17; I² = 54; Figure 7H), movement VAS at 24 h (MD = 0.02, 95% CI: −0.29 to 0.32; I² = 13%; Figure 7I), movement VAS at 48 h (MD = −0.50, 95% CI: −1.04 to 0.04; I² = 61%; Figure 7J), morphine consumption in second 24 h (MD = −0.80, 95% CI: −1.92 to 0.33; I² = 0%; Figure 7K), and number of people using additional analgesics (RR = 1.28, 95% CI: 0.71 to 2.31; I² = 0%; Figure 7L).

Discussion

The findings of this meta-analysis provide compelling evidence for the role of PIBI in CRA following thoracic surgery. By synthesizing data from multiple studies, we have demonstrated that PIBI significantly reduces LAs consumption and provided a wider sensory blockade compared to CI. These results underscore the potential of PIBI as a more effective infusion mode for CRA following thoracoscopic surgery. Existing meta-analyses have focused on comparing CI and PIBI in labour analgesia and upper or lower limb surgeries.^12,25 However, these analyses may not be directly applicable to thoracoscopic surgery due to differences in surgical technique, pain mechanisms, and patient characteristics. We consider that the different mechanisms between PIBI and CI is the main reason for the difference in results.

CRA have long been utilized to provide sustained analgesia by delivering LAs via catheters. Traditionally, CI has been the standard method, where a constant rate of LAs is administered to maintain a stable analgesic effect.²³ The mechanism of CI is based on maintaining a stable and continuous blockade of nerve conduction.²³ PIBI is a technique that involves the periodic administration of a bolus dose of LAs, allowing for rapid increases in LAs concentration at the nerve site during each bolus administration.²³ This can result in a more extensive and even coverage of the nerve fibers, enhancing the overall efficacy of the block.²³ Based on the above evidences, it is not difficult to explain why the sensory blockade range of PIBI is wider than CI.

Additionally, the popular CRA include TPVB, ESPB, and SAPB in thoracic surgery. The primary reasons for their popularity are the demonstrated effectiveness following a single injection, as well as the ease and safety of the ultrasound-guided puncture and catheterization techniques. For TPVB, the LAs diffuse within the thoracic paravertebral space, effectively blocking the ventral and dorsal rami of the spinal nerves, as well as the sympathetic chain.²⁶ For ESPB, the LAs are injected into the fascial plane between the erector spinae muscle and the transverse processes of the vertebrae, allowing the LAs to spread and block the branches of the spinal nerves.²⁷ For SAPB, the LAs is injected into the fascial plane either superficial or deep to the serratus anterior muscle, blocking the lateral cutaneous branches of the intercostal nerves, which innervate the anterolateral and lateral aspects of the thoracic wall.²⁸ Based on the evidence from the previous paragraph, the paravertebral space and these fascial planes have a more significant obstruction to the diffusion of LAs for CI. Additionally, the difference of LAs consumption and postoperative pain scores show that the efficacy attenuation rate of PIBI is slightly better than that of CI.

By administering the LAs in intermittent doses, PIBI reduces the continuous systemic absorption of the drug.²⁹ However, PIBI cannot avoid fluctuations in LAs concentration and may lead to side effects associated with high plasma levels of LAs.²⁹ Conversely, CI allows for a more predictable pharmacodynamic profile, as the LAs concentration remains relatively constant.²³ This can be advantageous in situations where a consistent level of analgesia is required over an extended period.²³ However, CI can lead to the accumulation of LAs in the tissues, potentially increasing the risk of systemic toxicity and side effects.²³ At present, the included studies do not reported any side effects and adverse reactions related to LAs, which only shows that these PIBI strategies are safe. Future studies should standardized strategy for PIBI to improve safety and reliability. Additionally, the movement VAS scores gradually increased between 24 and 48 hours, indicating that the severity of inflammatory pain was gradually intensifying, and regional analgesic techniques did not significantly alleviate inflammatory pain.

Limitations

Despite the promising results, this meta-analysis has several limitations that warrant consideration. First, the heterogeneity among studies, particularly in terms of catheter placement techniques, concentration and dosage of LAs, postoperative analgesia protocols, and patient populations, may have influenced the pooled estimates. Future studies should adopt standardized definitions for CRA and report outcomes in a uniform manner to improve comparability. Second, the majority of included studies were conducted in pulmonary surgery, limiting the generalizability of our findings to esophagus, mediastinum, and pleura surgeries. Further research is needed to evaluate the feasibility and effectiveness of two infusion strategies in diverse clinical environments. Another critical limitation is the lack of long-term follow-up data on chronic pain outcomes. Post-thoracotomy chronic pain is a significant complication, and CRA may play a role in its prevention. However, none of the included studies assessed the incidence of chronic pain beyond the immediate postoperative period. Future research should incorporate long-term follow-up to determine whether PIBI or CI reduces the risk of chronic pain and improves quality of life in thoracic surgery patients. Finally, the optimal concentration and dosage of LAs for PIBI remains uncertain. While existing studies have not reported the occurrence of side effects and adverse reactions related to LAs, concerns about safety may limit the clinical application of PIBI. Pharmacodynamic and pharmacokinetic evaluations are needed to determine the optimal infusion regimen for PIBI.

Conclusions

In conclusion, there was no significant clinical differences between PIBI and CI for resting VAS at 6 h. Meanwhile, the same results were observed in other pain scores, opioid consumptions, and number of people using additional analgesics. However, PIBI provided a lower LAs consumption and a wider sensory blockade compared to CI for CRA following thoracoscopic surgery. Further research is required to address the current gaps in the literature, including the need for standardized protocols, long-term follow-up, and pharmacodynamic and pharmacokinetic evaluations. As the field of regional anesthesia continues to evolve, CRA has the potential to become a cornerstone of postoperative pain management in thoracic surgery, ultimately enhancing patient care and recovery.