Percutaneous gastrojejunostomy in amyotrophic lateral sclerosis

We have performed a retrospective review of the use of a percutaneous gastrojejunostomy in patients with amyotrophic lateral sclerosis (ALS). Forty-one patients with initial bulbar manifestations of ALS and 32 patients with initial limb manifestations underwent a percutaneous gastrojejunostomy under fluoroscopic control using the Rankin gastrojejunostomy tube. Survival characteristics were compared with 86 bulbar onsetting and 207 limb onsetting ALS patients who did not require nutritional support. The 30-day mortality rate was 9.6% (respiratory death in three bulbar onsetting patients and four limb onsetting patients) and the 30 day morbidity rate was 4.1% (one operative site infection and intraperitoneal leakage in two patients). The most frequent long-term complication was the requirement for tube changing (blockage in six; dislodgment in two). Gastric reflux was not described amongst the treated patients. Overall survivorship (symptom onset to death) was less in the bulbar onsetting patients receiving a gastrojejunostomy tube than in the control population (median survival 22.0 vs. 33.7 months, respectively, P=0.005). As a group, the median survivorship for limb onsetting patients was not different for those receiving a gastrojejunostomy than for those who did not. However, a significant reduction in survival was observed in limb onsetting patients receiving a gastrojejunostomy early in the course of their disease (P=0.001) compared to those with a longer duration prior to the procedure. This was not observed in the bulbar onsetting patients. In both patient populations, no relationship was observed between survival post-gastrojejunostomy and the severity of pulmonary involvement at the time of the intervention, serum chloride, or age at onset. These studies demonstrate that a percutaneous gastrojejunostomy is a well-tolerated and safe alternative technique for enteral nutritional support in ALS patients. It also offers the advantage of not requiring either a general anaesthetic at the time of the procedure or instrumentation through the oropharynx. We have also observed that limb onsetting patients requiring a gastrojejunostomy early in the course of their illness are in a distinctive, less favorable, prognostic group.