Coeliac plexus block in the management of chronic abdominal pain due to severe diabetic gastroparesis

Background

Gastroparesis is defined as abnormal gastric motility characterised by delayed gastric emptying in the absence of mechanical obstruction. Symptoms include nausea, vomiting, early satiety, abdominal pain, and weight loss. The pathogenesis of pain in these patients is poorly understood and most treatments have been largely unsatisfactory.1 In this report, we describe a patient with chronic abdominal pain and malnutrition due to severe diabetic gastroparesis treated successfully with coeliac plexus block (CPB).

Case presentation

In January 2009, a 50-year-old woman with a 20 year history of type 1 diabetes complicated by retinopathy, nephropathy, neuropathy, and severe gastroparesis status following placements of a percutaneous endoscopy gastrostomy (PEG) tube (for venting) and a percutaneous endoscopy jejunostomy (PEJ) tube (for feeding), presented to our institution with chronic, persistent, and severe abdominal pain, recurrent nausea and vomiting, and severe malnutrition. The patient’s past medical history was significant for a failed pancreas transplant in 2004 and pancreatectomy of native pancreas in 2007 for recurrent pancreatitis. Notably, she also previously underwent pyloroplasty and gastric electrical stimulator (GES) implantation for her gastrointestinal symptoms without significant improvement; the implant was subsequently removed. Gastric scintigraphy in 2007 revealed notably delayed gastric and small bowel transit, and grade 3 (severe) gastroparesis (fig 1). Outpatient management of her abdominal pain had consisted primarily of opiate analgesics and gabapentin. Notably, her abdominal pain at times was complicated by constipation due to narcotic overuse. At the time of her presentation to our institution, the patient received most if not all of her nutrition through the PEJ tube, since oral feedings usually resulted in severe abdominal distention and pain that were only partially relieved with decompression through her PEG tube. However, the patient had recently discontinued PEJ tube feedings because of sharp abdominal pain that occurred during the feedings and reflux of contents into the stomach. Consequently, she experienced an unintentional decline in her body mass index from an already low 17 kg/m² to 15 kg/m² within the 2 weeks before admission.

Physical examination revealed a distention of the abdomen and diminished bowel sounds. Contrast enhanced computed tomography (CT) of the abdomen and pelvis showed pronounced stomach distention, with minimal contrast advancing beyond jejunum; however, there was no evidence of obstruction. The PEG tube was placed to intermittent suction for decompression and she was allowed no liquids or food by mouth. Intravenous fluids were used to maintain hydration. Barium contrast injected through the PEJ tube flowed freely, but very slowly, into non-dilated loops of jejunum with a small amount of reflux into the proximal region. In an attempt to resume enteral feedings and to avoid total parenteral nutrition, PEJ tube feedings with Peptamen were initiated at a rate of only 10 ml/h. Nevertheless, the patient’s hospital course was complicated by unrelenting and increasing abdominal pain with increases in the rate of PEJ tube feedings (to 30 ml/h). However, there was no evidence for obstruction or reflux of tube feedings into the stomach; tube feeding residual volumes and the amounts of fluid vented through the PEG tube were nil. Her pain did not improve with gabapentin, acetaminophen, or non-steroidal anti-inflammatory drugs and was only partially relieved with opiate analgesics. However, opiates precipitated and worsened her constipation.

Therefore, we considered CPB as an alternative means of pain management. Using a retrocrural approach under fluoroscopic guidance, CPB was performed by injecting local anaesthetic and steroid bilaterally anterior to the L1 vertebral body (fig 2). Pain control was adequately achieved based on a mean decline in numeric pain intensity scores from 8 (out of 10 on a 10 point scale) down to 4 after the procedure; pain control was maintained throughout her hospitalisation. Adjuvant analgesic therapies were also used including scheduled gabapentin, acetaminophen (paracetamol) as needed, a heating pad, and a lidocaine patch. Intravenous narcotics, a fentanyl patch, and tramadol were weaned and eventually discontinued with improvement in the regularity of her bowel movements. The patient continued to endorse minimal abdominal discomfort and tolerated advancement of PEJ tube feedings to a goal rate of 60 ml/h and resumption of oral fluid at an intake of 750 ml/day before discharge from the hospital. Symptoms of nausea were controlled with a combination of ondansetron and prochlorperazine administered by way of the PEJ tube as needed. The patient’s glucose concentrations were adequately managed with a combination of NPH insulin and Aspart.

Outcome and follow-up

The patient was evaluated by her primary care physician as an outpatient 14 weeks following CPB. She continued to endorse minimal abdominal discomfort, was weaned off her gabapentin, had successfully avoided any narcotic use, and was able to increase her PEJ tube feedings to 72 ml/h. With her symptoms of abdominal bloating, nausea, and vomiting under control, the patient was able to advance her oral intake from fluids to a pureed diet.

Discussion

The pathogenesis of delayed emptying in patients with longstanding type 1 diabetes is poorly understood. Gastric and small bowel motility mechanisms are deranged, largely due autonomic neuropathy, and reductions in the number of intrinsic inhibitory neurons and pacemaker cells (interstitial cells of Cajal) involved in motor coordination.2^,3 Abdominal pain has been recognised as a disabling symptom in many individuals with gastroparesis.4 To date, there are no data from controlled trials to guide the choice of agents for pain management in patients with gastroparesis. Agents commonly used in clinical practice include tricyclic antidepressants, gabapentin, and pregabalin with varying results.5 Pain in gastroparesis has been postulated to stem from sensory rather than motor dysfunction, which might explain why prokinetic medications improve predominant symptoms of nausea, vomiting and bloating, but do not seem to relieve abdominal pain. Treatments aimed at reducing sensory afferent dysfunction may be more effective.6

CPB has long been used to provide analgesia for upper abdominal pain. The celiac plexus contains visceral afferent fibres that innervate the abdominal viscera from the distal oesophagus to the transverse colon. From this plexus, afferent visceral nociceptive signals return to the spinal cord along the splanchnic nerves.7 A number of studies have demonstrated the efficacy of coeliac plexus neurolysis and CPB for reducing pain and opiate requirements in individuals with pain from chronic pancreatitis or pancreatic cancer.8^,9 To our knowledge, there have been no reports on the role of CPB in managing pain due to severe gastroparesis. In this case, successful analgesia was achieved and maintained for 10 weeks following CPB in our patient. This intervention allowed discontinuation of all opiate requirements, with noticeable improvement of narcotic induced constipation. Individuals with severe gastroparesis require supplementation with enteral tube feedings if they have experienced unintentional loss of 10% or more of usual body weight within the previous 3 to 6 month period, repeated hospitalisations for refractory symptoms or hydration, interference with delivery of nutrients and medications, and ongoing nausea and vomiting.10 Adequate pain control in our patient allowed us to resume PEJ tube feedings successfully and therefore prevented us from pursuing an alternative course of long term parenteral nutrition and its high cost and risks of complications.