Continuous surgical multi-level extrapleural block for video-assisted thoracoscopic surgery: a retrospective study assessing its efficacy as pain relief following lobectomy and wedge resection

Introduction

Minimally invasive thoracic surgery by means of video-assisted thoracoscopic surgery (VATS) has been shown to cause less postoperative pain than thoracotomy^1–4. Even though it is not as painful, VATS still requires adequate postoperative analgesia. As part of a multimodal analgesic regimen, different techniques using local anaesthetics can be used to block pain impulses from the surgical area. Epidural, paravertebral and intercostal blocks are all possible modalities. Compared to epidural block, using continuous extrapleural, paravertebral or intercostal block is thought to offer a more limited thoracic block, with possibly fewer side-effects.

For thoracotomy, several prior studies, including a Cochrane systematic review from 2016, all conclude that epidural and paravertebral block offer similar pain relief, with a paravertebral block possibly having fewer complications^5–8. The benefit vs. risk for the paravertebral technique has, however, been argued⁹.

The optimal pain management approach following elective VATS is still not known. A review by Steinthorsdottir et al.¹⁰ could not provide any firm recommendation. No firm conclusion could be drawn assessing available evidence around thoracic epidural, multilevel and single paravertebral, paravertebral catheter, intercostal catheter, interpleural infusion and long thoracic nerve block. The most recent study comparing epidural and percutaneous paravertebral block by Kosiński et al.¹¹ showed that the paravertebral continuous block technique was a feasible and safe alternative to epidural analgesia. Hutchins et al.¹² found ultrasound-guided continuous paravertebral catheter to provide prolonged pain control and superior patient satisfaction compared with single-shot intercostal block after VATS.

In 2014, continuous extrapleural block after VATS was implemented as a standard technique for pain management following VATS surgery at Karolinska University Hospital.

The aim of the study was to assess the quality of pain treatment after VATS with the routine use of a multi-level continuous extrapleural block as part of a multimodal analgesic strategy.

Methods

Results

A total of 510 patients received an extrapleural catheter during the period studied.

A thoracotomy was performed as the primary procedure in 26 patients. The remaining 484 patients were scheduled for VATS procedures. In 30 cases VATS was converted to thoracotomy. Neither patients that underwent primary thoracotomy nor patients converted to thoracotomy are included in our analysis. Of 454 cases analysed, 150 procedures were classified as wedge resection, 264 cases as lobectomy and the remaining 40 cases as miscellaneous (thymus resection, extirpation of a nodulus, lymph node(s), cyst or hamartoma and decortication with or without wedge resection) (Figure 1).

Figure 1. Patient inclusion flow chart.

Baseline characteristics are presented in Table 1. All patients were followed in-hospital from day of operation (POD 0) until POD 3, except for two patients discharged on POD 1, and 37 patients discharged on POD 2, reducing the total number of pain assessments. Missing data occurred where we could not retrieve a pain score or opioid dose, either due to patient discharge or lacking data in the charts. We lack NRS scores with increasing frequency from POD 0 to POD 3.

Pain at rest and during movement after VATS

At rest, pain was rated as mild, with median NRS rated 3-3-1-1 for POD 0 to 3. Assessed during movement, pain was rated moderate during POD 0 and 1 and mild the remaining days (median NRS 4-4-3-3 for POD 0 to 3). The proportion of patients exhibiting mild pain at rest increased from 55% on POD 0 to 81% on POD 3. The percentage of patients experiencing severe pain at rest decreased from 15% to 6% (Figure 2).

Figure 2. Pain after VATS for day of operation through postoperative day (POD) 3, categorized in mild, moderate and severe pain.

Pain at rest (a) and in movement (b). Percentage of patients with NRS scores available. Absolute numbers are shown in the centre of bars. Patient discharge and missing data cause diminishing absolute numbers during progression of postoperative course.

Pain after VATS wedge resection compared to VATS lobectomy

Pain at rest after VATS wedge resection was reported as median NRS 2-2-1-1 compared to pain after lobectomy with median NRS 3-3-2-2 for POD 0 to 3. At rest, more patients were experiencing mild pain and fewer patients were experiencing severe pain after wedge resection than lobectomy all postoperative days except POD 2 where an equal percentage of patients were experiencing mild, moderate and severe pain (Figure 3a, b). The pain pattern, mild pain versus non-mild pain, showed a difference in pain at rest for wedge resection vs. lobectomy at POD 1 (p = 0.03) and POD 3 (p = 0.006), respectively.

Figure 3. Pain at rest after VATS wedge resection (a) and lobectomy (b) for day of operation through postoperative day (POD) 3, categorized in mild, moderate and severe pain.

Percentage of patients with NRS scores available. Absolute numbers are shown in the centre of bars.

For POD 0 to 3 median NRS reported in movement was 3-4-3-2 after wedge resection and NRS 5-5-3-3 after lobectomy (p = 0.05 for POD 0).

Opioid consumption

Median oxycodone consumption was 10 mg per day for POD 1 to 3 (range: 0 to 210 mg, one patient with preoperative high dose opioid treatment). Median cumulative oxycodone consumption for POD 1 to 3 was 35 mg (range: 0 to 600 mg).

After VATS, opioid treatment was not required by 25% of patients at POD 1, 28% at POD 2 and 20% at POD 3. During the first three postoperative days in total, 15% of patients did not require oxycodone at all (Figure 4).

Figure 4. Number and percentage of patients requiring no opioid, low-dose opioid and high-dose opioid for postoperative day (POD) 1 to 3 and for total cumulative duration of POD 1 to 3.

Absolute number shown inside bars. Categorized data, for definition of low and high opioid dose, see the “Statistics” section of the Methods. Cumulative opioid dose is calculated only where data is complete for all days POD 1 to 3.

Comparing opioid consumption after wedge resection and lobectomy

The percentage of patients not requiring opioid POD 1 to 3 was minimally higher after VATS wedge resection. Conversely, the percentage of patients with a high opioid dose was larger for all postoperative days after lobectomy (Figure 5a, b). Patients not needing any opioid during any of POD 1–3 was 14% for both wedge resection and lobectomy. We did not find a significant difference in oxycodone dose after wedge resection and lobectomy for POD 1 to 3 or cumulative for POD 1 to 3.

Figure 5. Number and percentage of patients requiring no opioid, low dose opioid and high dose opioid for postoperative day (POD) 1 to 3 and for total cumulative duration of POD 1 to 3 after video-assisted thoracoscopic surgery wedge resection (a) respectively lobectomy (b).

Absolute number inside bars. Categorized data, for definition of low and high opioid dose see text under “statistics”. Cumulative opioid dose is calculated only where data is complete for all days POD 1 to 3.

Discussion

In this retrospective study we found that a surgeon placing an extrapleural catheter with continuous infusion of levobupivacaine 2.7 mg/ml at a rate of 5 ml/h was a valuable part of our multimodal pain treatment. Pain was well controlled, pain was at rest overall reported as mild and rated moderate initially and mild from POD 2 during movement. We found VATS lobectomy to be more painful than VATS wedge resection.

To achieve this level of analgesia, a median dose of only 10 mg oxycodone per day was needed. However, only 15% of our patients did not need any oxycodone at all and 47% needed more than a low dose of oxycodone during the postoperative course. This indicates that there is a potential to further improve the non-opioid part of our multimodal analgesic strategy. We did, however, see a relatively low incidence of opioid-related adverse effects, PONV was experienced in only 16% of patients.

It should be acknowledged that we used a fixed continuous infusion of 13.5 mg levobupivacaine per hour after an initial bolus of 75 mg regardless of age, body weight and surgery. To our knowledge, the optimal dose for a continuous extrapleural (or paravertebral) block is not known. A higher dose of local anaesthetic, either through higher concentration or higher rate of infusion, might offer superior analgesia. Simultaneously, the risk for adverse events due to toxicity might increase. This dose was chosen taking recommended maximal daily doses into account. Single-shot paravertebral with ropivacaine and adrenaline has been shown to decrease the systemic uptake of ropivacaine¹⁵. Addition of adrenaline might allow for more local anaesthetics to be infused. Kosiński et al.¹¹ observed that better analgesia was achieved using a mixture of bupivacaine and adrenaline when infused through a continuous paravertebral block rather than continuous epidural, even implying an additive effect of adrenaline itself. Future trials should address systemic absorption and toxicity for continuous blocks, measuring plasma concentrations of the local anaesthetic used.

Another possibility to increase the analgesic effect of our extrapleural catheter could be to add adjuncts other than adrenaline to the local anaesthetic infused. In a prospective study, Xu et al.¹⁶ showed that adding dexmedetomidine to a single-shot paravertebral block with ropivacaine resulted in better analgesia from 8 to 48 hours after VATS. For patients with unilateral multiple rib fractures, Mohta et al.¹⁷ showed equal pain relief with a lower dose of ropivacaine when a continuous paravertebral block contained a low dose of fentanyl in addition to ropivacaine and adrenaline. At the same time, Bauer et al.¹⁸ did not show an additional analgesic effect of sufentanil added to ropivacaine in a continuous paravertebral block. Whether a surgical extrapleural continuous block in VATS combining local anaesthetics, adrenaline and a short-acting opioid is superior to a block with only local anaesthetics is, to our knowledge, unknown.

We used levobupivacaine for our block. It is possible that alternative local anaesthetics offer better analgesia. Lidocaine may be an alternative having both a local block effect as well as systemic anti-inflammatory and possibly analgesic properties^19–21.

This study shows also that different procedures with a difference in trauma applied cause a varying intensity in pain. However, we did not find a significant difference in opioid requirement for different procedures, even though fewer patients needed a high oxycodone dose after VATS wedge resection. Still, tailored analgesia with different types of regional analgesia or different contents infused may be the way to offer patients the optimal balance between analgesia and side-effects, which may improve postoperative recovery.

There are weaknesses in our study. This is a retrospective patient-record-based study, and all patients operated during the period were included. The wide range of oxycodone dose (0-210 mg/day) may be a result of our retrospective approach. Patients with a preoperative history of chronic pain and/or ongoing opioid treatment were not specifically identified. We used 5-mg increments in oral oxycodone dose. Using a patient-controlled analgesia regimen might have shown a more accurate opioid requirement. It should also be acknowledged that several patients were discharged before POD 3. It is not unlikely that patients discharged earlier and lost to follow-up could have experienced the least pain and least need for opioid treatment. This might skew our results towards higher median pain and higher opioid requirement.

Conclusion

We found that a continuous surgeon-placed extrapleural catheter block to be a valuable and seemingly safe addition to our multimodal procedure specific analgesia after VATS. Whether the efficacy of the block can be improved by increasing local anaesthetic and/or adding adjuncts warrants further investigation.

Data availability

Dataset 1. Raw patient data assessed in the present study. Data include pain scores at postoperative days (POD) 0–3 at rest and with movement, alongside basic demographic information. DOI: https://doi.org/10.5256/f1000research.16857.d224364²².