DISCUSSION
In our retrospective study, anemia was noted in approximately 92% of patients presenting with AGI bleeding at discharge and treatment for anemia was planned at discharge in only 10.2% of discharged patients and the large remaining patient group did not receive any treatment for anemia. It was also previously shown in the study of Bager et al.10 that 84% of patients presenting to the hospital with non-variceal upper AGI bleeding and discharged from intensive care unit in Denmark also had anemia during discharge and only 16% of patients received iron therapy. Because there are no guidelines about iron therapy that the patients followed with the diagnosis of AGI bleeding and with anemia will receive following discharge, the rate of patients receving iron therapy is rather low though the approach of clinicians is variable. Therefore, guidelines are needed on this subject and long term studies should be carried out about the course of anemia with long term iron therapy in these patients.
Similarly, anemia was noted in 91,3% of our patients and only 15 patients received iron therapy and further investigations were not carried out regarding anemia. This shows that clinicians do not need further investigations associating GI bleeding with IDA. But this will lead to failure to treat other causes of anemia in case of overlap of IDA with conditions like chronic disease anemia or B12 / folic acid deficiency. Another point is that if the patient does not have severe anemia, the clinician may not take anemia seriously and plan further investigation regarding anemia. It is important to investigate iron deficiency in patients with AGI bleeding, because it has been shown in the literature that in these patients a mean increase of 2 g/dL in Hb levels has been observed with iron therapy in a 4-week period following discharge 11. In our study also, a mean increase of 2.96 g/dL has been observed in Hb levels of the patients given oral treatment in a three-month period. Only 1 patient of the 3 patients given IV treatment came to control visits and it was seen that there was 1 gr/dL increase in Hb level.
In our study, a significant difference was not found between the rates of anemia in patients receiving (63.6%) and not receiving (83.3%) iron therapy after a 3-month follow up (p>0.05). However, mean hemoglobin level of patients receiving iron therapy when anemia was noted at discharge was found to be higher than that of patients not receiving therapy and this difference was statistically significant (p<0.05). In conclusion, iron therapy was shown to increase Hb levels and there may be many reasons for the lack of a significant difference between anemia rates of groups receiving and not receiving iron therapy. A follow up period shorter than 6 months being insufficient for filling iron depots is an important factor. Continuation of chronic low blood loss due to the persistence of the gastrointestinal lesion in the patient was also considered.
In addition, another factor for the failure of anemia to improve may be the low compliance of patients particularly to oral iron therapy. Selection of the type of iron therapy should depend on the clinical condition of the patient (primary GIS disorder, pathophysiology, inflammation, presence of co-morbidities and malabsorption). Timely and regular iron therapy will also reduce the need for blood tranfusions of the patients. In the case of persistence of anemia, it should also be considered that the lesion causing GIS bleeding has not fully healed during follow up and still causes blood loss.
Only approximately 1/3 of our patients followed and discharged with the diagnosis of AGI bleeding came to control visits. Being uninformed about the condition of the patients who are lost to follow up is a major healthcare problem. In addition, anemia status of the patients before AGI bleeding is not fully known because this is a retrospective study. It cannot be predicted how this can affect mean Hb levels of the patients at discharge. Nevertheless, long term anemia status of the patients could not be evaluated because we evaluated 3-month follow up data of the patients after discharge.
In conclusion, anemia is seen at a very high rate like 92% at discharge in the patients presenting with AGI bleeding in our retrospective study and most patients are discharged without receiving a prescription for the treatment of anemia. Control presentation during follow up is inadequate and anemia persists in 80% of patients coming to follow up visits. Studies with larger number of patients about the treatment and follow up to be conducted and guidelines providing consensus on this subject are needed.