The reliability of the blood urea nitrogen/creatinine (B/C) ratio in localizing the bleeding site in pediatric patients with gastrointestinal hemorrhages was evaluated retrospectively. Supplementary experiments on rats and rabbits were also conducted.
The study population included 69 children ranging in age from 6 days to 16 years (mean age, 7y8m) with succesful radiographic, endoscopic, and surgical identification of the bleeding site. Patients with renal disease, liver disease and coagulopathy were excluded.
The upper gastrointestinal bleeding site was segregated from the lower one defined by the ligament of Treitz.
For the 38 upper gastrointestinal bleeders, B/C values ranged from 7.8-69.0 (mean 28.8.±15.4 SD). For the 31 lower gastrointestinal bleeders the values ranged from 4.3-28.5 (mean 18.9±6.9 SD). A significant difference between the two groups was noted after (p<0.01) analysis. No patients with lower gastrointestinal bleeding had values higher than 30. In the group of 17 bleeders with a value higher than 30, 8 patients were examined again from 2-5 days after they had been identified as having bleeding to prove that the bleeding had stopped. At the time of identification of no-bleeding, B/C values were calculated at under 30 in every patient. There was no significant correlation between Hb (g/d
l) and B/C.
With respect to our experiments on animals, the B/C ratio was evaluated in rats with gastric or small intestinal ulcers. Though B/C was 34.4±4.4 under normal conditions, it increased significantly to 114.0 ±8.8 in rats with gastrointestinal ulcers, while in those with small intestinal ulcers B/C did not increase significantly (40.3 ±12.3).
Twenty m
l of blood from ear veins of rabbits was infused into a) stomach b) small intestine (no-hemolysis blood) c) small intestine (hemolysis blood).
For a) and c), B/C increased in a similar pattern. For b) the increase was small. Infusing 5 m
l, 10 m
l, and 15 m
l of the whole blood into the stomach indicated that the volume of blood did not influence the change of B/C values, while the infusion of 3 m
l of serum caused only a small increase.
According to these studies the increase in the B/C ratio after upper gastrointestinal hemorrhage has been postulated to result from the procedure in which the hemorrhage blood is fragmented or hemolysed by HCI and digestive enzymes in the gastric juices; and consequently the BUN in red blood cells as well as in serum is rapidly absorbed from the upper small intestine.
Volume of blood has little influence upon B/C. Calculation of the B/C ratio in the initial evaluation of gastrointestinal bleeding may prove useful in the following situations.
1) When B/C is higher than 30, upper gastrointestinal bleeding is strongly suspected.
2) Reducing of B/C>30 to values under 30 indicates thatmacroscopic bleeding has been stopped.
View full abstract