Statistics from Altmetric.com
A recent article,1 a commentary,2 and two letters3,4 in Archives have revealed controversy over the place of partial splenectomy in portal hypertension in cystic fibrosis (CF). We wish to contribute to the debate with a case report:
Our male patient was homozygous for the ΔF508 mutation. He was pancreatic insufficient, his lungs were colonised with Pseudomonas aeruginosa from an early age, and he had two episodes of allergic bronchopulmonary aspergillosis. When he was 8 years old, abdominal ultrasound showed variable echogenicity of the liver compatible with cirrhosis with thick bile in the biliary tree. Treatment with ursodeoxycholic acid was commenced. Recurrent abdominal pain associated with severe gastro-oesophageal reflux led to an anti-reflux procedure being performed when he was 9 years old. A gastrostomy button was placed at the same time for night time supplementary feeding. Cirrhosis of the liver was confirmed intraoperatively. Over the next few years a massive splenomegaly developed. Full blood count showed features of hypersplenism but he remained asymptomatic with respect to the haematological abnormality. At the age of 13 years he developed severe abdominal pain in the area over the spleen. Oral analgesia was not sufficient to deal with this ongoing pain and he was unable to attend school, exercise, or do chest physiotherapy over a number of months. He had two episodes of probable melaena. He developed severe, intercurrent shoulder tip pain secondary to diaphragmatic irritation from splenic infarcts. Computerised tomography of the abdomen showed the spleen’s span to be 30 cm, with two infarcts. Opiates were given to control pain but it proved to be intractable in an otherwise stoical patient. Eventually, because of the risk to his lungs, his poor quality of life and the risk posed to his gastrostomy by the massive spleen, partial splenectomy and possible splenorenal shunting were planned. Pneumococcal vaccine was prescribed. His white cell count (WCC) was 1.5×109/l, platelet count 58×109/l, and INR 1.6. At laparotomy, perisplenitis in the diaphragmatic area necessitated a total splenectomy. Shunting was not undertaken. The spleen weighed 1834 g and there were numerous infarcts. Postoperatively he did well, patient controlled analgesia being used to encourage early mobilisation. Eight days later elective banding of oesophageal varices took place. Follow up endoscopy showed that this had ablated all the vessels. Two years later he no longer has abdominal pain, has not had severe infections, has a normal full blood count (WCC 12.3×109/l, haemoglobin 141 g/l, platelets 486×109/l), and has stable lung function.
The debate on the justification for removing all or part of the spleen in patients with CF and portal hypertension hinges on two considerations: indications and risks In their commentary, Kelly and de Ville de Goyet2 emphasised the risks: infection, compromising future transplantation, while questioning the indications in the cases presented by Thalhammer et al: hypersplenism and discomfort.1 In their rebuttal, Thalhammer and colleagues3 emphasise the hypersplenism and not the pain and discomfort described in their case reports. In their accompanying letter, Chazalette and colleagues4 do not mention pain as an indication. We would agree with Kelly and de Ville de Goyet2 that hypersplenism in the absence of significant consequences is not on its own an indication for this major procedure (we note the number of re-laparotomies required in these small series) but would emphasise that quality of life and local effects of the size of the spleen may justify the surgical and immunological risks.