The role of prophylactic nasogastric tube decompression in patients undergoing abdominal surgery as seen at Muhimbili Medical Centre

Study the importance of inserting or not inserting a nasogastric (Ryle's) tubes (NGT) prophylactically in patients undergoing abdominal surgery. All patients undergoing elective and emergency abdominal surgery in surgical wards at Muhimbili Medical Centre, with either biliary, gastric, duodenal, intestinal, peritoneal and other intra-abdominal surgical conditions, between February 1999 and November 1999, were prospectively randomised to one of the following groups: group I: those in whom the nasogastric tube was retained after surgery (controls); group II: those in whom the nasogastric tube was removed three to six hours postoperatively. The patients were monitored for the time of resumption of bowel movement, abdominal distension and vomiting, anastomotic leakage, wound infection, wound dehiscence, pneumonia, atelectasis, discomfort, length of hospital stay and deaths. Exclusion criteria included those who had oesophagectomy and those who were unconscious at the time of admission. Bowel movement was considered active when the patient passed flatus/stool or both and had no abdominal distension or vomiting. These were the indications to stop intravenous fluids, start ambulation and consider discharge home. Nasogastric tube re-insertion was indicated in those who developed gross abdominal distension or vomiting more than three times in group II. Two hundred and forty consecutive patients were studied, 120 patients in group I (86 males and 34 females, mean age 36.78 years [range 12-76 years]), and 120 patients in group II (100 males and 20 females, mean age 38.96 years [range 12-83 years]). Both group I and II patients were similar in age, case distribution and type of surgery. One hundred and fifteen patients (95.8%) were treated successfully without NGT decompression (group II). In group II patients, there were three (2.5%) deaths (one due to septicaemia, the second because of cardiac arrest and the third due to metastatic gastric malignancy),and one case of anastomotic leak (0.8%). There were no cases of pneumonia, wound dehiscence, wound infection, or delay in return of gastrointestinal function, but two patients required re-insertion of the NGT. In the control group (group I), there were nine (7.5%) deaths, (three from severe haemorrhage, another three because of septicaemia, one dead of severe acute pancreatitis, one because of cardiac arrest, and one because of severe peritonitis and history of local herbs intoxication), sixty six patients (55%) had discomfort due to the NGT, three (2.5%) wound dehiscence, one patient each (0.8%) had wound infection and anastomotic leakage and no