Electrolyte pattern in patients with dynamic intestinal obstructions as seen in Mulago Hospital, Kampala, Uganda

Abstract

BACKGROUND: Electrolytes are important in the management of patients with intestinal obstruction. They may not be investigated in an emergency situation. Therefore an expected pattern may be useful in the management of intestinal obstruction patients. METHODS: Patients reporting to mulago hospital with dynamic intestinal obstruction were evaluated in a prospective study. Before any resuscitation venous blood was analyzed for sodium, potassium, chloride and bicarbonate. Relation of the level of each of these to the duration of obstruction, occurrence of vomiting, location of obstruction and presence of strangulated intestines was analyzed. Univariate analysis was performed to identify risk factors for developing different electrolyte levels. RESULTS: The 82 patients included 57 (69.5%) with ileal obstruction and 25 (30.5%) with clonic obstruction. In ileal obstruction, the different levels of sodium, potassium and chloride depended on the duration of obstruction with stastical significance. The p-values were <0.001, 0.020 and 0.036 respectively. But the bicarbonate was not (P-values of 0.252) In colonic obstruction, the same electrolyte trend was seen but with a 1 ½ days longer mean duration of symptoms. However, time did not affect potassium and bicarbonate levels significantly in colonic obstruction patients. The p-values were 0.900 and 0.499 respectively. The p-values for sodium and chloride were 0.038 and 0.013 respectively. Vomiting did not influence the level of sodium, potassium, chloride or bicarbonate, p-values being 0.328, 0.701, 0.320 and 0.459 respectively. Sodium deficit was preset in 93.0% of the patients. Anion gap levels were not significantly influenced by time, p-values being 0.474. However, an anion gap less than 10 was present in 42 patients (53.2%). The base deficit was also not influenced by time. The p-value was 0.431. When time was not controlled for, 65 (82.3%) patients had a base deficit of more then 2.5 mEq/l. Location of obstruction did not influence serum sodium, potassium and bicarbonate. The p-value were 0.062, 0.107 and 0.364 respectively. But serum chloride was significantly affected by location irrespective of duration. A higher percentage of patients with chlonic obstruction 14(56.0%) had had hypochloraemia than ileal obstructed patients 17 (29.8%). There was no patient with colonic obstruction and serum chloride levels above normal. The p-value was 0.049. Stranulation of intestines significantly rises the occurrence of hyperkalaemia was a p-value of 0.042. CONCLUSION: Na+, K+ CI- levels are influenced by duration of the obstruction and location of the obstruction. But bicarbonate levels were not. Strangulation of bowel was difficult to determine basing on abdominal tenderness. Strangulation of bowel influences potassium levels only. Vomiting does not influence Na+, K+, CI- and HCO3- levels.