In patients with intra-abdominal malignancies, chronic abdominal pain (CAP) has a prevalence up to 50% [26]. Furthermore, patients with uncontrolled background pain may develop breakthrough cancer pain in up to 70% of cases, with huge impact on patients’ quality of life and disability [3].

Due to its pain transmission the coeliac plexus is a veritable target for controlling pain from upper abdominal organs. As part of a multimodal approach, CPN is a commonly used interventional pain management strategy to decrease pain, increase quality of life and reduce opioid therapy in patients with intra-abdominal malignancies [26, 30].

Back in 1979, Ward and colleagues [28] defined the coeliac artery as the most reliable landmark for locating the coeliac plexus and it has been reported that the injection site of the neurolytic solution should be cephalad to the coeliac trunk [5, 6]. Given the variations in patients’ anatomy, possible anatomic alterations by tumour masses or previous surgery or radiation therapy, clearly visible needle tip and contrast spread, CT-guidance offers several advantages. Different techniques (single needle or bilateral injections), approaches (anterior para-aortic [11, 19, 22], bilateral posterior para-aortic [2, 26], posterior transaortic [11, 13], trans-intervertebral disc [10, 26]) and patient positions (prone or supine [26]) have been described, with no clear superiority for optimal results in any of these techniques.

However, De Cicco et al. [5] reported that injections cephalad of the coeliac trunk should be performed to obtain wider spread of the injected solution. Furthermore, they divided the coeliac area in the frontal plane into four quadrants defined by a horizontal line passing caudad to the root of the coeliac trunk and by a vertical line in the midline of the ventral wall of the aorta. They showed that only complete spread of the neurolytic in the upper and lower right and left quadrants seems to guarantee optimal and long-lasting pain control. They found that when the neurolytic solution spreads to only one quadrant, poor pain relief should be expected and when fewer than 4 quadrants are reached by the neurolytic only a small percentage of patients will experience adequate pain relief.

Although the goal of CPN is complete spread of the injectate to achieve long-lasting analgesia [5], irregular spread of neurolytic solution may occur because of regional infiltration by tumour masses, or anatomic alterations by previous radiation therapy and/or surgery. In a retrospective evaluation [6] of patients whose coeliac area was infiltrated by tumour masses or distorted by previous radiation therapy or surgery, over 90% of patients showed completely hampered spread of the injectate. In these patients one can expect poor or even no pain relief when the injectate reaches only parts of the coeliac area.

We performed a CT-guided anterior approach with a targeted needle tip position in the midline just anterior to the ventral wall of the aorta cephalad to the root of the coeliac trunk.

Similar to De Cicco et al. [5] the coeliac area was divided, in the frontal plane, into 9 almost equal quadrants, with the origin of the coeliac trunk as the central structure (see Fig. 1). Although we found that spread of the neurolytic solution bilaterally of the aorta leads to improved pain relief compared to unilateral right contrast spread, we did not find any other correlation between pain relief and spread of neurolytic solution in the coeliac plexus. Furthermore, we did not observe any association between the quadrants of contrast spread and pain relief, neither in the number of quadrants reached by the neurolytic solution, nor in any particular quadrant with contrast or different combinations of quadrants with neurolytic solution. Also, in patients with complete spread of the contrast medium in the coeliac plexus there was only a trend for slightly better pain relief compared with partial spread of the neurolytic in the coeliac area.

In our evaluation we found hampered spread of the neurolytic solution due to tumour masses in 13.3% of cases. Interestingly, there was a weak significant correlation between better pain relief in patients and hampered spread of contrast medium.

Surprisingly, based on the spread of contrast medium in the coeliac area, patterns of injectate spread, and calculated area of injectate, the assessors could not correctly anticipate the pain relief or post-procedural NRS, nor expect the duration of pain relief after CPN. In addition, there was only a statistical trend for decreased pain relief with increased distance of the needle tip to the coeliac trunk.

Our findings stand in contrast to the findings of De Cicco et al. [5, 6] who reported that only complete neurolytic spread in the coeliac area can guarantee long-lasting analgesia [5] and secondly that the decision to perform CPN must be based on the anatomic conditions of the coeliac area in each patient [6].

A possible explanation is that less advanced cancer infiltration, which indeed is connected with pain, but is mostly visceral without a multifactorial component, may be more easily suppressed by CPN [23] and thus incomplete neurolytic spread in the coeliac area or even hampered spread of the neurolytic solution could lead to good and long-lasting pain control.

On the other hand, pain evolution is unpredictable [16]. If other areas of neural or somatic structures are involved, the efficacy of the CPN may be decreased as the intervention aims to block the sympathetic pathways, rather than somatic afferents [16, 17]. Furthermore, pancreatic cancer-related pain is a complex condition involving many different pathophysiological mechanisms [1]: tumour location [4], autonomic plexus invasion [7], locoregional and distant tumour spread [20], malignant obstruction and intraluminal activation of pancreatic enzymes [15], small bowel distension [14], perineural tumour invasion of intrapancreatic nerves and neurogenic inflammation [12, 25].

These findings lead to the hypothesis that it is not essential to have the perfect sickle-shaped spread of neurolytic solution in CPN in palliative patients with intra-abdominal malignancies for adequate pain control. In this respect, when needle positioning is difficult, hampered spread of the neurolytic or incomplete or partial spread of the neurolytic in the coeliac area is found, clinicians may relinquish a further attempt at needle positioning or complete spread of contrast medium. Furthermore, we found that it is not essential to place the needle tip as close as possible to the coeliac trunk to achieve good pain relief.

We speculate that these findings will be a relief to both, clinicians and patients, because it does not seem not to be necessary to make multiple attempts at complete spread of contrast medium in the coeliac area for adequate pain relief. It may be possible to shorten the duration of the procedure and so increase patient acceptance and satisfaction due to a more rapid and less afflicted intervention. Furthermore, as good pain relief can be achieved form a safe distance to the coeliac trunk, the possible risk of bleeding complications may be further reduced.

If irregular spread of neurolytic solution occurs, one cannot predict the direction of the injectate spread in the coeliac area. It is possible that different pressure areas, textures and grain may prevent or promote the spread of the injectate [6]. Interestingly, there was no correlation between the needle tip position and spread of neurolytic solution in any anatomical direction or contact between the neurolytic solution and intra-abdominal organs respectively.

One typical side effect of CPN is back pain, which mostly radiates to the shoulder, resulting from neurolysis of sensory nerve fibres [11]. One can also expect transient pain at the injection site, or diarrhoea and hypotension due to sympathetic blockade [2]. Thrombosis of the coeliac trunk or vasospasm of the coeliac trunk leading to hepatic, splenic, gastric, or bowel infarction [8, 26] are rare complications, as are major bleeding [29], retroperitoneal haematoma [18], or lower extremity paralysis [24]. Complications due to poor needle placement, e.g. kidney injuries or neurological complications due to inadvertent injection of neurolytic agent are scarce due to CT-guidance [11].

Given the spread of contrast medium and contact between the neurolytic and different intra-abdominal organs or plexuses we tried to predict different kinds of side effects of CPN, as described above. Although the liver, stomach and adrenal glands with the associated plexuses where most likely reached by the contrast medium, no special complications or side effects occurred. We found that “other” plexuses were less often reached by the neurolytic when the distance of the needle tip from the coeliac trunk increased.

We mainly expected intestinal hypomotility or alteration of stomach peristalsis as possible side effects of our CPN. Although we found a 10% higher rate of correctly predicted complications compared to false predictions, the statistical difference was not significance and it was not possible for assessors to predict different side effects correctly.

There are several important limitations in our study that deserve mention. This study was retrospectively performed in a single centre and used a real-world clinical practice model. Therefore, different pain physicians planned the pain management strategies. The final decision to conduct CPN could have been influenced by their own preconceptions and may have biased the study population. One potential limitation was that after the intervention, no CT scan of the complete coeliac area was performed. So, spread of the cranial and caudal injectate could not be evaluated. It remains unclear, whether the furthermost cranial or caudal spread of the neurolytic in the coeliac area is correlated with pain outcomes. Here further studies with post-interventional CT scans are needed, although their significance in clinical routine is questionable and would increase the radiation exposure of these patients.

Despite these limitations, we feel that our large study population, our standardized procedure as well as the separate review of these blinded CT scans by an anaesthetist and pain specialist, an anatomist, and a radiologist allow us to reach valid clinically relevant conclusions.