Acute lower gastrointestinal bleeding (ALGIB) is usually defined as overt bleeding from the GI tract distal to the ligament of Treitz leading to hospitalization 1–4. The epidemiology and time trends of upper GI bleeding have been widely investigated, whereas very limited data exist on the incidence and outcomes of lower gastrointestinal bleeding (LGIB).
The incidence rate for ALGIB has been reported to be 20.5/100 000 1 in a population believed to resemble that of San Diego in the USA in the early 1990s and 33/100 000 2 in a population considered to be representative of the population of Spain in 2005, with the incidence increasing with older age in both studies. Good-quality epidemiological studies on LGIB are lacking, but it was recently reported that upper GI bleeding events seem to be decreasing whereas LGIB events seem to be increasing 2. Thus, prospective studies on the incidence, etiology, and prognosis of well-characterized patients with LGIB are required.
The role of drugs as risk factors for ALGIB has not been investigated thoroughly outside clinical trials. Several studies have shown that NSAIDs seem to increase the risk of LGIB events as they do in the upper GI tract 5,6, although other investigations did not show this relationship 7–9. Most of these studies lack a control group of nonbleeders. The role of anticoagulation drugs such as warfarin, low-dose aspirin, platelet inhibitor drugs, as well as selective serotonin receptor inhibitors (SSRIs) and bisphosphonates in the context of LGIB is not clear 9–11. A recent retrospective study from Sweden did not find evidence for the impact of increased use of drugs that potentially promote bleeding on the incidence and fatal outcome of bleeding, although the severity of GI bleeding has increased over the years 12. In Iceland, every citizen has a specific personal identification number and, since 2006, a registry of individual drug dispensation has been in use, providing the opportunity to link patients with this database, which includes all prescribed drugs in the country from 2002 onwards 13.
The aims of the current study were to study the incidence of ALGIB in a population-based setting, with a particular focus on the role of drugs in its etiology and, furthermore, to study the etiology of overt LGIB and analyze the overall outcomes.
Materials and methods
Population and case finding
The study was prospective and included all patients who underwent upper GI endoscopy and/or colonoscopy in the National University Hospital of Iceland from 1 January to 31 December 2010. A control group of nonbleeding patients undergoing endoscopic examinations during the same time period was matched with the patients for sex and age (±5 years). Definitions were as follows.
Acute lower gastrointestinal bleeding
- Passage of bright red blood per rectum or maroon colored without hematemesis leading to hospitalization or occurring in a hospitalized patient.
- Melena with no signs of bleeding in upper GI endoscopy along with hospitalization or occurring in a hospitalized patient.
Of all those who underwent colonoscopy, many individuals were excluded; they can be divided into five groups:
- Individuals undergoing colonoscopy for reasons other than LGIB or a suspicion of LGIB.
- Patients suspected of occult bleeding because of anemia, iron deficiency anemia, and/or a positive hemoccult test with no diagnosis at endoscopy.
- Patients with melena and no signs of bleeding in upper GI endoscopy and colonoscopy.
- Individuals who had overt GI bleeding but were not hospitalized.
- Patients with occult bleeding with a confirmed cause of bleeding.
To calculate the incidence of ALGIB, every individual who did not have a registered home in the greater metropolitan area of Reykjavík was excluded, as the National University Hospital of Iceland is bound to serve that population. Information about the population was provided by the Office for National Statistics in Iceland 14.
Before the endoscopic procedure, the gastroenterologist noted the indication for the endoscopy and its findings were recorded and later verified by the examination of individual medical reports. The gastroenterologists recorded whether or not GI bleeding was suspected and/or present and whether or not the bleeding was clinically significant. Bleeding was considered clinically significant if the patients required blood transfusions (Hb<100 g/l), became hemodynamically unstable, were admitted to the ICU, required surgery, or died. Other variables noted were a history of GI bleeding, signs of bleeding or blood in the colon during colonoscopy, and the extent and findings of the colonoscopy. Nurses in the endoscopy ward interviewed the patients thoroughly before the endoscopy in terms of their drug-use history. The drugs recorded were the following: NSAIDs, low-dose aspirin, warfarin, SSRIs, bisphosphonate drugs, platelet inhibitors, corticosteroids, heparin/low-molecular-weight heparin, and proton pump inhibitors.
The use of these drugs had to be on a regular basis and not as on-demand therapy. NSAID users were defined as patients who used NSAIDs daily for a minimum of 5 days. In many patients, corticosteroids and heparin/low-molecular-weight heparin were started in hospital and the use of these drugs had to be started before the first endoscopic examination.
Laboratory values were collected and were as follows: hemoglobin (g/l), platelet counts, creatinine (µmol/l), prothrombin time (s), and INR value. Data on blood transfusions were provided by the Blood Bank of Reykjavík, which provides blood products to the National University Hospital. All information on transfusions is stored in an electronic information system (Prosang, Databyran, Sweden) and full traceability is ensured.
To explore the role of drugs in the etiology of LGIB, all patients with LGIB were matched for sex and age (±5 years) with a control group including those individuals who underwent colonoscopy in the same institution during the same time period without being suspected of or having LGIB. Controls were matched with the bleeders after collection of data from all these patients. We were able to match patients one to one and therefore there was no selection of nonbleeding patients matched.
The Icelandic Medicines Registry records all prescriptions issued outside of hospitals and nursing homes in Iceland and it is run by the Directorate of Health in Iceland. It has been operated since 1 January 2006, but contains data from 2002 onwards; about 2 300 000 registrations are added to the database annually. To further improve the reliability of the drug history, records from the Icelandic Medicines Registry were examined. These records contained information on every prescription issued for the individuals participating in the study from the 1st of January 2009 until the 31th of December 2010. Patients who were started on the predetermined drugs in hospital were connected to a pharmaceutical database in the National University Hospital (Theriak, London, UK).
Every individual medical record was reviewed and all data were verified.
This study was approved by the Data Protection Authority of Iceland and the Bioethics Committee of Iceland.
All data were processed in Microsoft Office Excel 2010 (Microsoft, Redmond, Washington, USA) and IBM SPSS Statistics (IBM, Armonk, New York, USA). The χ2 goodness-of-fit test and the Fisher Exact test were used to test differences between groups in dichotomous variables. The unpaired Mann–Whitney U-test was used to compare continuous variables. Variables with a significant P-value in the univariate analysis were entered in multiple logistic regression analyses in an attempt to identify independent predictors of having a LGIB and clinically significant bleeding. All tests were two-tailed and were carried out at a 5% significance level. The results are presented as medians and interquartile range or means and SD.
Individuals 18 years of age and older in the catchment area of the National University Hospital of Iceland were 151 008 at the 1st of January 2010. Thus, the crude annual incidence during the year 2010 was 87/100 000 inhabitants. Age-standardized incidence rates are presented in Table 1. The highest incidence rates were found in the oldest age groups (Table 1).
Figure 1 shows a flow chart of the patients included in the study and the reasons for exclusions in various parts of the study. During the 12-month study period, a total of 370 patients had LGIB. A total of 163 (44.1%) fulfilled the predetermined criteria for ALGIB, 163 (44.1%) had overt bleeding but were not hospitalized, and 44 (11.8%) had occult LGIB. Overall, 200 patients were hospitalized but seven individuals with occult LGIB were not hospitalized. About one-third of the cohort had a previous history of GI bleeding and 81% had anemia; further information on laboratory parameters can be found in Supplementary Table 1.
The primary indications for the endoscopies were blood in feces and/or bleeding per rectum (n=155, 95.1%) or melena (n=8, 4.9%), with or without concomitant anemia. Many patients had more than one indication. The diagnoses in the total study cohort are shown in Table 2. The most common diagnosis was bleeding from diverticulosis, followed by ischemic colitis, and approximately similar proportions were associated with inflammatory bowel disease, colorectal cancer, and hemorrhoids (Table 2). In 9.2% of the cases, the bleeding was because of an unknown cause and various other uncommon etiologies were also found (Table 2).
The overall cohort had a median age of 68 (range 18–90) and 50.3% were women; the controls had a median age of 69 (range 18–90) and 49.7% were women (NS for both comparisons). Comparison of the use of potentially harmful drugs leading to ALGIB showed that a significantly higher proportion of patients with LGIB had regular use of NSAIDs, low-dose aspirin, and low-molecular heparin compared with nonbleeding controls (Table 3). None of the other type of drugs differed significantly between patients with and without ALGIB (Table 3). In a logistic regression analysis, use of the following drugs was independently related to bleeders: NSAIDs [odds ratio (OR) 2.569, 95% confidence interval (CI) 1.296–5.092, P=0.007], low-dose aspirin (OR 1.675, 95% CI 1.023–2.743, P=0.04), and heparin/LMWH (OR 5.125, 95% CI 1.441–18.23, P=0.012).
Patients who had clinically significant bleeding were older (Table 4). The vast majority of cases with colorectal cancer had clinically significant bleeding and all patients with bleeding because of vascular ectasia had clinically significant bleeding (Table 4). Patients with ischemic colitis rarely had clinically significant bleeding (Table 4). Taking into consideration the use of drugs, patients on warfarin and those on the combination of warfarin and low-dose aspirin significantly more often had a clinically significant bleeding than those who were not treated with these drugs (Table 4). In a stepwise multiple logistic regression with clinically significant bleeding as the dependent variable, the following variables were found to be independent predictors of clinically significant bleeding: age (OR 1.021, 95% CI 1.003–1.039, P=0.023), ischemic colitis (OR 0.158, 95% CI 0.056–0.444, P<0.001), nonhemorrhoidal bleeding (OR 6.959, 95% CI 1.85–26.169, P=0.004), and the use of both warfarin and low-dose aspirin (OR 9.478, 95% CI 1.102–81.501, P=0.041).
A total of 75% of patients were diagnosed at the first colonoscopy and 20% at the second one. Only 12 patients (7%) required endoscopic hemostasis and vascular ectasia was the cause of bleeding in five (42%) of those 12 patients; no individual experienced rebleeding. Further data on colonoscopies and endoscopic treatment can be found in Supplementary Tables 2 and 3, respectively.
Approximately 40% of patients received blood transfusions and most patients received 2–3 U of blood with a median value of 3 U (range 0–18 U). Of the 64 individuals receiving blood transfusion, 21 (32.8%) had diverticulosis as a cause of bleeding, 14% had colon cancer, and only 6% had ischemic colitis. Only two (1.2%) patients died directly because of ALGIB. One patient had prostate cancer that formed a colonic fistula resulting in uncontrollable bleeding and the other patient had ALGIB of an unknown cause along with end-stage chronic obstructive pulmonary disease.
In this prospective population-based study, the incidence of ALGIB is, to our knowledge, the highest reported to date. Interestingly, the incidence of ALGIB and acute upper gastrointestinal bleeding (AUGIB) was the same (Fig. 1). This is in contrast to a previous series showing upper bleeding to be much more common than lower bleeding from the GI tract. The most common causes of ALGIB were diverticulosis and ischemic colitis. Mortality during hospitalization was very low and only associated with severe comorbidity. NSAIDs, low-dose aspirin, and low-molecular heparin seem to increase the risk of LGIB, whereas other drugs do not seem to be of major importance.
GI bleeding is a major health issue in most parts of the world. Both increasing age of the population in most Westernized countries and increased use of drugs interfering with blood coagulation make GI bleeding an important health issue. The incidence of ALGIB has been reported to be from 21 to 43/100 000 inhabitants/year 1,2,12; therefore, the incidence of 87/100 000 is certainly the highest reported in the literature to date. One of the reasons for the high incidence of ALGIB in the current study is probably the fact that this was a prospective study. All patients who underwent endoscopy during the year 2010 were enrolled in a systematic manner and the indications registered as well as outcome by experienced gastroenterologists. Thus, it is unlikely that we missed any potential case with either AUGIB or ALGIB who required hospitalization. Previous studies on the incidence of ALGIB have been retrospective and patients were identified on the basis of a search for diagnostic codes 1,2,12, which might lead to underestimation of the incidence. Recently, a study from the Netherlands has reported that the positive predictive value of a broad search strategy on the basis of diagnostic codes to identify patients with upper GI bleeding is very low 15. Another part of the explanation might be that during the past decades, there has been a change in the overall type of GI bleeding. Previous studies have found upper GI bleeding to be much more prevalent than LGIB 2,16,17. A recent multicenter study from Spain found a clear decreasing trend in the upper GI bleeding episodes during the period of 1996–2005, with a concomitant increase in LGIB 2. If it is assumed that the trends of increased lower and decreased upper GI bleeding would continue, that might in part explain the high incidence of ALGIB and the fact that in the current study, the incidence of ALGIB and AUGIB is the same.
The etiologies found in the current study are quite similar to the largest studies that have been carried out (Table 5). Also, of all the colonoscopies, 9% yielded no result, which is similar to the range of 4–17% with an unclear diagnosis found in other studies (Table 5).
A major focus of this study was to analyze the role of drugs in the etiology of ALGIB. Compared with a control group who underwent colonoscopy for nonbleeding symptoms, the drugs that were more commonly associated with ALGIB were NSAIDs, low-dose aspirin, and low-molecular-weight heparin. The association between the use of NSAIDs and LGIB found in the current study is in agreement with most studies that have examined this relationship 6,12,16,19–25. Indeed, in most studies, the risk of bleeding with NSAIDs was similar in upper and lower GI bleeding 5,6,19,21. In the current study, nonbleeding controls had a significantly lower use of NSAIDs than those with ALGIB. Thus, the results of the current study, together with those of the previous studies outlined above, show without much doubt that NSAIDs are clear risk factors for ALGIB. Much less data are available for low-dose aspirin and the risk for LGIB. Interestingly, a significantly higher proportion of our patients were on low-dose aspirin [acetylsalicylic acid (ASA)], 37 versus 25% among bleeders than the nonbleeding controls, respectively. Arroja et al.3 reported the use of platelet-aggregating drugs in 35.4% of patients with ALGIB, which is very similar to our results, but this study did not either include a control group without bleeding. Thus, our results add to the existing literature further supporting the role of ASA in LGIB. It is difficult to establish a causal relationship between low-dose aspirin and ALGIB in these patients and our results only allow us to conclude that the use of this drug was associated with bleeding. Chan et al.26 have recently reported that ASA users with a history of LGIB, who continued low-dose aspirin, had more than a two-fold increased risk of recurrent bleeding compared with those who discontinued ASA. These results, together with the association between ASA and ALGIB in our patients, suggest that these drugs are probably important in the etiology of the bleeding. Interestingly, the combination of warfarin and low-dose aspirin was found to be an independent predictor of clinically significant bleeding.
The current study was not sufficiently powered to carry out reliable subanalyses among the studied drugs. The current study was also underpowered to assess a potential relationship between bisphosphonates and LGIB. However, no tendency was observed in the few patients who were on these drugs and bleeding and there was no evidence of an increased risk of upper GI bleeding among bisphosphonate users in another study 11. The use of low-molecular heparin was significantly higher in bleeders than in nonbleeders. Arroja et al. 3 found that heparin use before bleeding was a risk factor for poor outcome. The use of low-molecular heparin and the relationship with LGIB requires further study.
The prognosis of the patients in the current study with ALGIB was generally favorable. The mortality from bleeding was found to be 1.2% and occurred in two patients and was associated with severe comorbidity. The overall mortality for ALGIB has been reported to range from 0% 3 to 7.2% 18. In most published series, the mortality related to bleeding ranges from 3 to 4% 4,27–29. Endoscopic therapy has been reported to be carried out in 3–22% of most published series (Table 5). Only 7.4% of patients were considered to require endoscopic therapy in the current study and this is only one-third of patients treated with endoscopic therapy in the multicenter but not population-based study from Portugal 3. Although the authors considered this to explain the low rate of urgent surgical intervention, which was carried out in 1.4%, the value of endoscopic therapy in these cases is unclear as the study was uncontrolled 3. No patient in the current study underwent urgent surgery. Our study was not aimed at identifying the risk factors for poor outcome and has insufficient power to show significant predictive parameters for the overall prognosis of these patients.
The incidence for ALGIB is the highest reported to date. The most common reasons for ALGIB were diverticulosis and ischemic colitis.
Mortality during hospitalization was very low and only associated with severe comorbidity. NSAIDs, low-dose aspirin, and low-molecular heparin seem to increase the risk for LGIB, whereas other drugs do not seem to be of major importance.
The authors deepest gratitude is extended to Magdalena Sigurðardóttir, RN, Elísabet Lilja Haraldsdóttir, secretary, and the doctors and nurses at the Endoscopic Department at the National University Hospital of Iceland.
A grant was received from the National University Hospital of Iceland Research Fund.
Conflicts of interest
There are no conflicts of interest.
1. Longstreth GF. Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: a population-based study. Am J Gastroenterol. 1997;92:419–424
2. Lanas A, Garcia-Rodriguez LA, Polo-Tomas M, Ponce M, Alonso-Abreu I, Perez-Aisa MA, et al. Time trends and impact of upper and lower gastrointestinal bleeding and perforation in clinical practice. Am J Gastroenterol. 2009;104:1633–1641
3. Arroja B, Cremers I, Ramos R, Cardoso C, Rego AC, Caldeira A, et al. Acute lower gastrointestinal bleeding management in Portugal: a multicentric prospective 1-year survey. Eur J Gastroenterol Hepatol. 2011;23:317–322
4. Makela JT, Kiviniemi H, Laitinen S, Kairaluoma MI. Diagnosis and treatment of acute lower gastrointestinal bleeding. Scand J Gastroenterol. 1993;28:1062–1066
5. Laine L, Smith R, Min K, Chen C, Dubois RW. Systematic review: the lower gastrointestinal adverse effects of non-steroidal anti-inflammatory drugs. Aliment Pharmacol Ther. 2006;24:751–767
6. Wilcox CM, Alexander LN, Cotsonis GA, Clark WS. Nonsteroidal
antiinflammatory drugs are associated with both upper and lower gastrointestinal bleeding. Dig Dis Sci. 1997;42:990–997
7. Doyle G, Furey S, Berlin R, Cooper S, Jayawardena S, Ashraf E, et al. Gastrointestinal safety and tolerance of ibuprofen at maximum over-the-counter dose. Aliment Pharmacol Ther. 1999;13:897–906
8. Rampton DS, Barton TP. Are prostaglandins cytoprotective in the human large intestine? The effect of indomethacin on rectal mucosal function and prostaglandin E2 release in vivo. Agents Actions. 1984;14:715–718
9. Wessinger S, Kaplan M, Choi L, Williams M, Lau C, Sharp L, et al. Increased use of selective serotonin reuptake inhibitors in patients admitted with gastrointestinal haemorrhage: a multicentre retrospective analysis. Aliment Pharmacol Ther. 2006;23:937–944
10. Khapra AP, Rose S. Drug injury in the upper gastrointestinal tract: effects of alendronate. Gastrointest Endosc Clin N Am. 2006;16:99–110
11. Etminan M, Levesque L, Fitzgerald JM, Brophy JM. Risk of upper gastrointestinal bleeding with oral bisphosphonates and non steroidal anti-inflammatory drugs: a case-control study. Aliment Pharmacol Ther. 2009;29:1188–1192
12. Ahsberg K, Hoglund P, Kim WH, von Holstein CS. Impact of aspirin, NSAIDs, warfarin, corticosteroids and SSRIs on the site and outcome of non-variceal upper and lower gastrointestinal bleeding. Scand J Gastroenterol. 2010;45:1404–1415
13. Landlæknir (the directorate of health in Iceland). Available at: http://www.landlaeknir.is/Pages/988
[Accessed 1 August 2011]
14. Hagstofa Íslands (Statistics Iceland). Available at: http://hagstofa.is/Hagtolur/Mannfjoldi
[Accessed 1 August 2011]
15. Valkhoff VE, Preciosa C, Mees M, Kuipers EJ, Sturkenboom MC, Trifiro G. Validation of the diagnosis for upper gastrointestinal bleeding in a Dutch medical record database. Gastroenterology. 2012;142:A-792
16. Peura DA, Lanza FL, Gostout CJ, Foutch PG. The American College of Gastroenterology Bleeding Registry: preliminary findings. Am J Gastroenterol. 1997;92:924–928
17. Lanas A, Perez-Aisa MA, Feu F, Ponce J, Saperas E, Santolaria S, et al. A nationwide study of mortality associated with hospital admission due to severe gastrointestinal events and those associated with nonsteroidal
antiinflammatory drug use. Am J Gastroenterol. 2005;100:1685–1693
18. Gayer C, Chino A, Lucas C, Tokioka S, Yamasaki T, Edelman DA, et al. Acute lower gastrointestinal bleeding in 1112 patients admitted to an urban emergency medical center. Surgery. 2009;146:600–606 discussion 606–607
19. Day JP, Lanas A, Rustagi P, Hirschowitz BI. Reversible prolonged skin bleeding time in acute gastrointestinal bleeding presumed due to NSAIDs. J Clin Gastroenterol. 1996;22:96–103
20. Holt S, Rigoglioso V, Sidhu M, Irshad M, Howden CW, Mainero M. Nonsteroidal
antiinflammatory drugs and lower gastrointestinal bleeding. Dig Dis Sci. 1993;38:1619–1623
21. Lanas A, Sekar MC, Hirschowitz BI. Objective evidence of aspirin use in both ulcer and nonulcer upper and lower gastrointestinal bleeding. Gastroenterology. 1992;103:862–869
22. Langman MJ, Morgan L, Worrall A. Use of anti-inflammatory drugs by patients admitted with small or large bowel perforations and haemorrhage. Br Med J (Clin Res Ed). 1985;290:347–349
23. Foutch PG. Diverticular bleeding: are nonsteroidal
anti-inflammatory drugs risk factors for hemorrhage and can colonoscopy predict outcome for patients? Am J Gastroenterol. 1995;90:1779–1784
24. Tsuruoka N, Iwakiri R, Hara M, Shirahama N, Sakata Y, Miyahara K, et al. NSAIDs are a significant risk factor for colonic diverticular hemorrhage in elder patients: evaluation by a case-control study. J Gastroenterol Hepatol. 2011;26:1047–1052
25. Shibuya T, Ohkusa T, Yokoyama T, Matsumoto K, Beppu K, Sakamoto N, et al. Colonic mucosal lesions associated with low-dose aspirin: a case control study. Scand J Gastroenterol. 2011;46:810–817
26. Chan FKL, Ki ELL, Tse YK, Au KWL, Ching J, Wong G, et al. Risk of recurrent lower gastrointestinal bleeding with continued low-dose aspirin therapy: a 10-year retrospective cohort study. Gastroenterology. 2012;142:A-132
27. Chaudhry V, Hyser MJ, Gracias VH, Gau FC. Colonoscopy: the initial test for acute lower gastrointestinal bleeding. Am Surg. 1998;64:723–728
28. Green BT, Rockey DC, Portwood G, Tarnasky PR, Guarisco S, Branch MS, et al. Urgent colonoscopy for evaluation and management of acute lower gastrointestinal hemorrhage: a randomized controlled trial. Am J Gastroenterol. 2005;100:2395–2402
29. Brackman MR, Gushchin VV, Smith L, Demory M, Kirkpatrick JR, Stahl T. Acute lower gastroenteric bleeding retrospective analysis (the ALGEBRA study): an analysis of the triage, management and outcomes of patients with acute lower gastrointestinal bleeding. Am Surg. 2003;69:145–149