RANDOMIZED CONTROLLED TRIAL
High-Dose Barium Impaction Therapy for the Recurrence
of Colonic Diverticular Bleeding
A Randomized Controlled Trial
Naoyoshi Nagata, MD,∗ Ryota Niikura, MD,∗ Takuro Shimbo, PhD,† Naoki Ishizuka, PhD,†
Kazuyoshi Yamano, MT,‡ Kyoko Mizuguchi, MSN,§ Junichi Akiyama, MD,∗ Mikio Yanase, MD,∗
Masashi Mizokami, PhD, and Naomi Uemura, PhD
Objective: We compared the clinical efficacy of barium therapy and conservative therapy in preventing recurrence in patients with diverticular bleeding.
Background: Previous case reports have indicated that barium impaction
therapy provides initial hemostasis for diverticular bleeding and prevention
against rebleeding.
Methods: After spontaneous cessation of bleeding, patients were randomly
assigned to conservative treatment (n = 27) or high-dose barium impaction
therapy (n = 27). Patients were followed up for 1 year after enrollment of the
last patient. The main outcome measure was rebleeding.
Results: Median follow-up period was 584.5 days. The probability of rebleeding at 30-day, 180-day, 1-year, and 2-year follow-up in all patients was 3.7%,
14.8%, 28.4%, and 32.7%, respectively. By group, probability at 1 year was
42.5% in the conservative group and 14.8% in the barium group (log-rank
test, P = 0.04). After adjustment for a history of hypertension, the hazard
ratio of rebleeding in the barium group was 0.34 (95% confidence interval,
0.12–0.98). No complications or laboratory abnormalities due to barium therapy were observed. Compared with the conservative group, the barium group
had significantly (P < 0.05) fewer hospitalizations per patient (1.7 vs 1.2),
units of blood transfused (1.9 vs 0.7), colonoscopies (1.4 times vs 1.1 times),
and hospital stay days (15 days vs 11 days) during the follow-up period. No
patients died and none required angiographic or surgical procedures in either
group.
Conclusions: High-dose barium impaction therapy was effective in the longterm prevention of recurrent bleeding, and reduced the frequency of rehospitalization and need for blood transfusion and colonoscopic examination.
ClinicalTrials.gov Identifier, UMIN 000002832.
Keywords: barium solution, colonic diverticular hemorrhage, long-term
follow up, natural history, preventive therapy
(Ann Surg 2015;261:269–275)
From the ∗ Department of Gastroenterology and Hepatology; †Clinical Research
and Informatics, International Clinical Research Center Research Institute;
‡Department of Radiology; §Department of Nursing, National Center for
Global Health and Medicine, Tokyo, Japan; and Department of Gastroenterology and Hepatology, National Center for Global Health and Medicine,
Kohnodai Hospital, Chiba, Japan.
Supported by a grant-in-aid from the Ministry of Health Labor and Welfare of Japan
and a grant-in-aid from the Ministry of Education, Culture, Sports, Science,
and Technology of Japan (271000) and the Grant of National Center for Global
Health and Medicine (22–302). The funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the article.
Disclosure: The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of
this article on the journal’s Web site (www.annalsofsurgery.com)
Reprints: Naoyoshi Nagata, MD, Department of Gastroenterology and Hepatology,
National Center for Global Health and Medicine, 1-21-1 Toyama, Shinjuku-ku,
Tokyo 162-8655, Japan. E-mail: [email protected].
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright ISSN: 0003-4932/14/26102-0269
DOI: 10.1097/SLA.0000000000000658
Annals of Surgery r Volume 261, Number 2, February 2015
A
cute lower gastrointestinal bleeding (LGIB) has an estimated
annual incidence rate of 20.5 patients/100,000,1 and is a common reason for hospitalization.1–3 The most common cause of LGIB
is colonic diverticular bleeding (CDB), which accounts for approximately 40% of cases.1
Approximately 76% to 92% of cases of CDB resolve with rest,
sedation, and blood transfusion.4–7 However, the rebleeding rate is
reported to be as high as 14% to 43%.4–9 Even when a definitive
diagnosis of CDB is made and therapeutic procedures are successful,
rebleeding still occurs in 11% to 38% of patients within 30 days after
endoscopic therapy,8,10,11 and in 30% after angiography.12 Recurrence
requires frequent examination, hospitalization, and blood transfusion,
and it decreases quality of life, and an effective means of prevention
is urgently required.
Among therapeutic procedures for CDB, a number of articles
have reported that high-dose barium enema use9,13–18 provides better
clinical outcomes for initial hemostasis and long-term prevention of
rebleeding than conservative therapy.9,15 To date, however, effective
preventive treatment against the recurrence of CDB has not been established. We hypothesized that barium impaction therapy has better
long-term efficacy in preventing recurrence than conservative therapy alone. Here, we conducted a randomized controlled trial with
extended follow-up to determine the efficacy and safety of barium
impaction therapy for CDB.
METHODS
Study Design
The study was conducted under a single-center, 2-arm, parallel group, open-label, randomized controlled design with a 1:1 treatment allocation, and included a minimum 1-year follow-up period
(ClinicalTrials.gov number, UMIN number: 000002832). The study
protocol was approved by the clinical research ethics committee of
the National Center for Global Health and Medicine (no. 765). All
patients provided written informed consent for participation in the
trial.
Setting and Patients
From December 2009 to December 2011, all patients admitted
to our hospital with painless overt signs of LGIB were treated jointly
by a team of gastroenterologists. Among these, patients meeting the
following inclusion criteria were eligible for entry in the trial: (1) age
between 20 and 90 years; (2) Japanese nationality; (3) spontaneous
cessation of bleeding after hospitalization; and (4) CDB. CDB was
defined as definite or presumptive on the basis of colonoscopy with
other imaging modalities. Definite diagnosis was based on colonoscopic visualization of colonic diverticulum, with active bleeding,
adherent clot, or visible vessel.19–21 Presumptive diagnosis was based
on fresh blood localized to colonic diverticula in the presence of a potential bleeding source on complete colonoscopy; or bright red blood
www.annalsofsurgery.com | 269
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 2, February 2015
Nagata et al
per rectum confirmed by objective color testing and colonoscopy that
demonstrated a single potential bleeding source in the colon, complemented by negative upper endoscopy or negative capsule endoscopy,
or negative nasogastric tube; or multidetector computed tomography
(MDCT) visualization of the extravasation of contrast medium localized into colonic diverticulum and colonoscopy showing a potential
bleeding site in an area of positive MDCT findings.19,20 Colonoscopy
was performed within 4 days of hospitalization. MDCT was performed within 12 hours of LGIB onset.
Patients were excluded if they presented any of the following:
(1) spontaneous cessation of bleeding within 4 days of hospitalization;
(2) shock requiring intensive medical care; (3) history of allergic
reaction to contrast dye; (4) intestinal obstruction; (5) history of
barium examination within 6 months; (6) difficulty in changing body
position; (7) pregnancy; (8) history of colonic resection; (9) diagnosis
of inflammatory bowel disease or colorectal cancer; or (10) coexisting
advanced cancer. Inclusion and exclusion criteria were established in
consideration of study safety and ethics, and data reliability.
Randomization and Masking
After hospitalization, all patients underwent fasting and rest
with drip infusion. Eligible patients were then assigned to one of
2 treatments: barium impaction therapy or conservative treatment
after spontaneous cessation of bleeding. Randomization was stratified
according to whether or not patients had a history of hypertension,
which is reportedly a common risk factor of CDB.22,23 A random
allocation sequence was then generated using a computer-generated
list of random numbers in blocks. To ensure concealed allocation,
investigators were informed of the assignment only after identification
of an eligible patient. These procedures ensured that investigators
were blinded to the allocation sequence.
Patients in the barium group underwent bowel preparation using 1 mL of sodium picosulfate hydrate and 250 mL of magnesium
citrate at least 4 hours before impaction therapy. Most also underwent
treatment with 2 L of polyethylene glycol solution the day before
treatment, during colonoscopic preparation. Barium sulfate was administered with an enema bag positioned 3 to 4 ft above the patient,
using a double-balloon enema tip. The final concentration of barium
was 200 w/v percentage (200 g/100 mL) with tap water to a total
volume of 400 mL. The rectal and anal cuffs were then inflated. All
patients were able to tolerate the treatment without incontinence. Barium impaction therapy was conducted by 2 gastroenterologists and
1 experienced radiological technician. After confirming the filling of
multiple colonic diverticula with the barium solution (Fig. 1), the enema tube was clamped to retain the solution in the entire colon, and
the patient was asked to frequently change position by rotating stepwise through the prone, left lateral, supine to right lateral positions
to ensure the filling of each diverticulum. Barium dosage, concentration, and administration method were conducted in the same manner
regardless of the bleeding location or the anatomical distribution of
diverticula. The enema tube was then withdrawn. Vital signs were
monitored throughout the procedure.
After randomization, all patients in the conservative group,
and in the barium group after impaction therapy, were started on a
liquid food diet, and gradually progressed to a solid diet over 3 days.
During hospitalization, blood transfusion was indicated in patients in
either group when the hemoglobin level fell below 7.0 g/dL, or below
8.0 g/dL in those showing unstable vital signs.
barium impaction therapy or endoscopic clipping, as determined on
a case-by-case basis.
Patients were followed up for 1 year after enrollment of the
last patient. After discharge, all patients were requested to visit the
hospital on a scheduled day within 1 month of discharge during
the observation period; every 3 months during the subsequent followup period; and additionally at any time in the case of bloody stools.
Patients who did not visit the hospital on a scheduled day were interviewed by telephone and encouraged to visit.
Outcomes
The primary outcome was recurrence of CDB. Patients who
experienced rebleeding after discharge continued to visit the hospital regularly during the follow-up period to identify further bleeding
episodes. Rebleeding was defined as significant amounts of fresh
bloody or wine-colored stools (>200 mL) without lower abdominal
pain, and was evaluated by either or both anoscopy and MDCT, within
12 hours of onset, wherever possible. Rebleeding prompted further
endoscopy whenever possible. Second-look endoscopy was not routinely performed when rebleeding occurred during hospitalization or
within 1 month of discharge, but was used to confirm rebleeding
and determine the need for intervention when frequent or massive
bleeding occurred.
Secondary outcomes were the number of hospitalizations,
blood transfusions, and colonoscopies within the observation period;
length of hospital stay; barium impaction therapy-related complications, including colon perforation and laceration of the rectal mucosa
by the enema tube13,24 ; and unfavorable changes in health, including
worsening of hemodynamic and respiratory status, development of
clinical symptoms, and abnormal laboratory findings.
Statistical Analysis
Sample size was calculated under the assumption that 43.4%
of the conservative group and 15.9% of the barium impaction therapy
group would experience recurrent bleeding.9 To detect a difference
between groups of at least 26% using a log-rank test with a 2-sided
alpha error of 0.05 and power of 0.80, entry period of 12 months,
and observation period of 12 months, we required a total of 29 events
and 24 patients in each treatment group. Baseline characteristics of
patients were compared using the Mann-Whitney U test, Pearson’s
χ 2 test, or Fisher exact test as appropriate, with recurrent bleeding as
the primary end point. The recurrence curve of CDB was estimated
using the Kaplan–Meier method, and a history of hypertension between groups was compared using the stratified log-rank test. Data
were censored at the time of the last visit or at the end of the followup period. Efficacy data were analyzed on an intention-to-treat basis.
Adjustment for hypertension, nonsteroidal anti-inflammatory drugs
(NSAIDs), and chronic renal failure, which are known to be associated with recurrence,7,25,26 was done using the Cox proportional
hazards model, and the proportional hazard assumption was assessed
by inspection of log-log plots.
The total number of rebleeding episodes per patient, hospitalizations, and need for colonoscopy within the follow-up period were
also compared between groups using the Mann-Whitney U test. All
statistical analyses were performed with Stata version 10 (StataCorp,
College Station, TX). All reported P values were 2-tailed, with a
P value of <0.05 considered statistically significant.
RESULTS
Measurements and Follow-up Procedures
Study Sample
After randomization, patients were asked to record the characteristics of their stools daily. Patients who experienced rebleeding
during protocol treatment received an alternative treatment, such as
Of the 71 patients admitted during the study period with painless LGIB, 61 were found to have CDB. After excluding 7 patients,
54 were recruited to the study (Fig. 2), of whom 27 underwent barium
270 | www.annalsofsurgery.com
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 2, February 2015
Barium Therapy for Diverticular Bleeding
FIGURE 1. X-ray findings of colonic diverticula. X-ray revealing multiple diverticula in the sigmoid colon (A), descending and
transverse colon (B), and ascending colon at (C) 1 day (D), 2 months (E), and 6 months (F) after barium impaction therapy.
impaction therapy and 27 conservative therapy. No patients were excluded from the intention-to-treat analysis.
Twenty-five patients (46%) had been previously diagnosed
with CDB (Table 1). Three patients were identified with SRH (stigmata of recent hemorrhage) on colonoscopy, giving a definitive diagnosis in 3 patients and a presumptive diagnosis in 51.
With regard to the severity of the initial bleeding, the 2 groups
were similar in the number of units of blood transfused before randomization and length of spontaneous cessation of bleeding (Table 1).
Other factors were also similar between the 2 groups, including obesity, alcohol and smoking consumption, urgent colonoscopy, anatomical characteristics of colonic diverticula, and CDB diagnosis data
(Table 1). Furthermore, the 2 groups were also similar with regard
to laboratory data, medication, and coexisting illness (Supplementary Table 1, Supplemental Digital Content 1, available at http://
links.lww.com/SLA/A546).
Primary Outcome
During hospitalization, further bleeding was documented in
4 patients (14.8%) who received conservative therapy alone, but in
none of those who received barium impaction therapy (Fisher exact
test, P = 0.11; Table 2).
After discharge, 12 patients in the conservative therapy group
and 5 in the barium group experienced rebleeding. Median follow-up
period was 584.5 days (range: 204–814). The probability of rebleeding at 30-day, 180-day, 1-year, and 2- year follow-up in all patients was
3.7% [95% confidence interval (CI), 0.91%–14.0%], 14.8% (95% CI,
7.70%–27.5%), 28.4% (95% CI, 18.2%–42.7%), and 32.7% (95% CI,
21.7%–47.4%), respectively (Fig. 3A). By group, the probability of
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
rebleeding at 1-year follow-up was 42.5% (95% CI, 26.1%–63.7%)
in the conservative group and 14.8% (95% CI, 5.83%–34.8%) in the
barium group (stratified log-rank test, P = 0.04; Fig. 3B). After adjustment for a history of hypertension, NSAIDs, and chronic renal
failure, the hazard ratio of rebleeding in the barium group was 0.34
(95% CI, 0.12–0.97).
Episodes of recurrent bleeding from colonic diverticula are
illustrated in Figure 2. The total number of rebleeding episodes within
the follow-up period was 25 (Table 2), with fewer episodes in the
barium group than in the conservative group (P = 0.06; Table 2).
Complications and Secondary Outcomes
No complications were observed during or after barium therapy, including colorectal perforation, diverticulitis, or laceration of the
rectal mucosa by the enema tube. In addition, no worsening of hemodynamic or respiratory status was observed. A number of clinical
symptoms were observed, namely nausea (n = 3), vomiting (n = 1),
abdominal distension (n = 4), and constipation (n = 5), but these
were mild to moderate and either resolved by dietary manipulation or
disappeared within 3 days. Constipation was resolved after treatment
with 1 mL of sodium picosulfate hydrate. No abnormal laboratory
findings were observed during hospitalization after barium impaction
therapy, such as a deterioration in WBC (white blood cells), CRP
(C-reactive protein), electrolytes, or creatinine levels.
During initial hospitalization, the total units of blood transfused
after randomization was smaller in the barium group than in the
conservative group, albeit without statistical significance (P = 0.23;
Table 2). Duration of the initial hospital stay did not differ significantly
between the 2 groups (P = 0.38; Table 2).
www.annalsofsurgery.com | 271
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 2, February 2015
Nagata et al
FIGURE 2. Flow chart of the study design. In the conservative group, 12 patients experienced rebleeding at 16, 51, 63, 96,
163, 183, 225, 284, 326, 334, 355, and 400 days, of whom 8 received conservative therapy, 2 underwent barium therapy,
and 2 underwent endoscopic clipping. Four of these 12 patients, namely 3 who received subsequent conservative therapy and
1 who received subsequent barium therapy, experienced the second rebleeding at 107, 195, 364, and 717 days. Of these
4 patients, 3 achieved hemostasis within the follow-up period while 1 experienced the third rebleeding at 411 days and received
conservative therapy. In the barium group, 5 patients experienced rebleeding at 30, 115, 131, 278, and 388 days, of whom 4
received conservative therapy and 1 underwent barium therapy. Three of these 5 patients experienced the second rebleeding
at 133, 370, and 392 days, respectively, with 2 going on to receive conservative therapy while 1 received barium therapy. ADL
indicates activities of daily living.
TABLE 1. Clinical Characteristics on Admission and at Follow-up∗
Baseline Characteristic
Male, n
Age, yr
BMI, kg
Waist circumference, cm
Smokers (never/occasional/daily), n
Smoking index†
Alcohol (non/light/ moderate/ heavy), n‡
Previous colonic diverticular bleeding, n
Distribution of colonic diverticula (right/left/bilateral colon), n§
Colonic diverticula (2–9/≥10 diverticula), n
Urgent colonoscopy, n¶
Identification of bleeding on colonoscopy, n¶
Units of blood transfused before randomization per patient, n
Length of spontaneous cessation of bleeding, d
Conservative Therapy
Group (n = 27)
Barium Therapy
Group (n = 27)
19
70 ± 11
24 ± 3
87 ± 8
14/8/5
806 ± 614
14/3/2/8
12
9/5/13
7/20
4
2
1.0 ± 1.9
3.1 ± 1.1
21
71 ± 13
23 ± 4
84 ± 11
11/9/7
890 ± 424
16/1/3/7
13
7/5/15
10/17
2
1
0.8 ± 1.9
3.0 ± 0.9
∗
Values presented with a plus/minus sign are means ± standard deviation. Some patients had more than one coexisting illness. The distribution and the number of colonic
diverticula were assessed by colonoscopy.
†Smoking index was evaluated for occasional and daily smokers, and defined as the number of cigarettes per day multiplied by the number of smoking years.
‡Alcohol consumption was divided into 4 groups: nondrinker, light drinker (1–180 g/wk), moderate drinker (181–360 g/wk), and heavy drinker (≥361 g/wk).
§Right-sided: transverse or proximal colon; left-sided: descending or distal colon; bilateral: around the entire colon.
¶Urgent colonoscopy was defined as that performed within 24 hours of admission. Colonoscopy was performed in 52 of the 54 patients in whom observation up to the
terminal ileum was possible. The remaining 2 were excluded because of a previous diagnosis of CDB by colonoscopy and exclusion of other sources of bleeding within
3 months of LGIB onset in one each.
BMI indicates body mass index.
272 | www.annalsofsurgery.com
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 2, February 2015
Barium Therapy for Diverticular Bleeding
TABLE 2. Outcomes After Randomization∗
Outcome
During initial hospitalization
Further bleeding
Units of blood transfused after randomization per patient, n†
Length of hospital stay per patient, d
Angiographic embolization or surgery, n
Follow-up period after discharge‡
Rebleeding episodes, n§
Episodes per patient, n†
Hospitalizations per patient, n
Units of blood transfused per patient, n
Length of hospital stay per patient, d
Colonoscopies per patient, n
Angiographic embolization or surgery, n
Conservative Therapy
Group (n = 27)
Barium Therapy
Group (n = 27)
4
1.1 ± 1.9
9.9 ± 4.2
0
0
0.4 ± 1.0
8.7 ± 2.8
0
17
0.6 ± 0.8
1.7 ± 0.7
1.9 ± 2.4
15 ± 8
1.4 ± 0.6
0
8
0.3 ± 0.7
1.2 ± 0.6
0.7 ± 1.2
11 ± 6
1.1 ± 0.3
0
P
0.11
0.23
0.38
0.06
<0.01
0.03
0.03
<0.01
∗
Plus/minus values represent means ± standard deviation.
†P value for comparison between groups was obtained using the Mann-Whitney U test.
‡Outcomes due to bleeding events including those that occurred during initial hospitalization. P value for the comparison between groups was obtained using the Mann-Whitney
U test.
§Total number of patients in the group who experienced recurrent bleeding during the observation period.
FIGURE 3. Probability of recurrent bleeding. (A) Incidence of rebleeding with a median follow-up period of 584.5 days.
(B) Probability of rebleeding was significantly higher in the barium group than in the conservative group (P = 0.04).
During the overall study period, in contrast, the total number
of hospitalizations, units of blood transfused, and colonoscopies per
patient was significantly lower in the barium group than in the conservative group (Table 2). Furthermore, the total length of hospital
stay per patient was also significantly lower in the barium group (P =
0.03; Table 2). No patients died or required angiographic embolization or surgical resection in either group within the follow-up period
(Table 2).
DISCUSSION
In this study, we evaluated the efficacy of high-dose barium
therapy against rebleeding in a series of patients with CDB. Most
cases were presumptive and had a history of CDB, and all cases
had spontaneous cessation of initial bleeding. The barium therapy
group showed a substantially lower incidence of rebleeding than the
conservative therapy group, with a hazard ratio of 0.34. Furthermore,
these patients had lower frequencies of rehospitalization, repeated
blood transfusion, and colonoscopic reexamination, and a shorter
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
length of hospitalization during long-term follow-up. These findings
suggest that high-dose barium therapy may be effective in this hitherto
difficult clinical condition.
The probability of rebleeding in this prospective cohort study
was higher than in Western populations. Previous studies have shown
low rebleeding ratios in Western populations,1,6 at 4% to 9% of patients at 1 year and 10% at 2 years, versus high ratios in Asians, at
20% to 35% at 1 year and 33% to 42% at 2 years.7,26,27 A rebleeding
rate of 27% at 1 month was recently reported in Japan.28 Our present
result is consistent with these Asian studies. The most likely explanation is related with differences in anatomical distribution patterns
of diverticula, which are predominantly left-sided in the West and
right-sided or bilateral in Asia.29
Several case reports or case series of barium impaction therapy
for severe CDB have been reported.9,13–18 Adams13 demonstrated for
the first time that 26 of the 28 massive bleeding episodes were resolved by barium impaction therapy, and that this treatment avoided
emergency colectomy in 3 patients. In addition, Koperna et al9
www.annalsofsurgery.com | 273
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 2, February 2015
Nagata et al
FIGURE 4. Flow chart for the management
of colonic diverticular bleeding. When patients are hospitalized for severe or continuous bleeding suspected from colonic diverticula, most cases resolve without procedures.
Urgent or elective colonoscopy with or without other modalities is indicated in accordance with the general condition of the patient as well as in consideration of institutional
and national background. However, stigmata
of hemorrhage is infrequently identified, and
thus many bleeding cases given a final diagnosis of diverticulosis will have a presumptive rather than definite diagnosis. Definite
diagnosis should be managed by endoscopic
therapy initially in the conventional manner,
possibly with the option of adding high-dose
barium impaction therapy in particularly patients with multiple diverticula, whereas presumptive cases have the additional option of
barium impaction therapy to prevent further
rebleeding.
reported failure rates of conservative therapy and barium therapy
with subsequent rebleeding of 43.4% and 15.9%, respectively,9 and
suggested that complications and mortality after surgery were higher
than those after barium therapy.9 The goal of intervention in CDB
is to identify SRH. However, the rate of SRH identification varies
widely, from 7.7% to 43% of LGIB cases,30–33 because the colon has
an anatomically large and complex surface area, often with multiple
potential sources, and because bleeding tends to be intermittent in
nature. Most diagnoses of CDB are therefore presumptive, accounting for as many as 87% of LGIB cases.1,19 Moreover, even when
SRH is identified and treated, some patients with multiple diverticula
will bleed again from another diverticulum.5 We consider that this
clinical picture supports the indication of barium impaction therapy
for presumptive CDB, particularly with multiple diverticula (Fig. 4).
Previous reports have also suggested that barium impaction therapy
has advantages in patients at risk for surgery or angiography; in patients with uncontrolled bleeding because of therapeutic procedures
or uncontrolled presumptive CDB; as well as in patients unable to
receive 4 to 6 L of bowel preparation or with multiple or inaccessible lesions.9,13–18 Concerns have been raised that retained barium
would preclude accurate colonoscopic visualization of rebleeding
cases. However, colonoscopy with full preparation in our present
cases did not interfere with visualization of the bowel lumen.
The fundamental mechanism of the effect of barium enema is
unknown, but 3 factors can be considered: (1) tamponade of the bleeding vessel through physical pressure caused by the barium solution,13
(2) a direct hemostatic effect of barium sulfate,13 and (3) protection
from intestinal fluids through the long-term presence of barium in the
diverticula. Barium enema administered from 3 ft (91 cm) above the
examination table produces an estimated intraluminal colonic pressure of 80 mm Hg, which is considered safe,34 whereas administration
from 6 ft (1.8 m) produces 140 to 168 mm Hg, which can result in
perforation.34 The source of CDB is thought to be arterial, which
would produce a mean arterial pressure of about 90 mm Hg.35 We
274 | www.annalsofsurgery.com
therefore administered barium from 3 to 4 ft above the table, equivalent to an intraluminal colonic pressure of 90 mm Hg, and this might
therefore have been effective in stopping the arterial bleeding. During barium coagulation, a tap water enema is reportedly better than
most barium suspensions as it contains no anticoagulants and is more
effective in clot formation.36 Moreover, use of a high concentration
of barium and retention in the colon may be related to increased viscosity and the facilitation of coagulation.16–18 Accordingly, we used
200 w/v percentage barium sulfate with tap water in the present study.
Finally, previous studies have shown that barium in the colonic diverticula or appendix can be retained without inflammation for several
months, and sometimes years,37,38 suggesting that barium impaction
is effective in preventing rebleeding as well as safe. Moreover, longterm impaction with barium may protect diverticula from the adverse
impact of colonic microflora in the development and progression of
diverticular disease.39
This study experienced no severe complications from barium
impaction therapy, suggesting the safety of barium enema of the
colon. Nevertheless, complications have been reported. These include perforation, trauma from the enema tip or retention balloon,
mucosal irritation, and thrombosis of the involved vessels.13,24 Of
these, colonic perforation is the most frequent, occurring in approximately 0.04% of patients.24 Contrasting with this, patients undergoing endoscopy are also at risk of perforation, particularly those with
colonic diverticula.40 Thus, both forms of CDB examination should
be carried out carefully.
This study has several limitations. First, it was conducted in
a non–double-blinded manner because of the difficulty in producing a suitable barium replacement. Although this might have biased the secondary outcomes, it is unlikely to have influenced rebleeding. Second, although we adjusted the multivariate analysis
for hypertension, NSAIDs, and chronic renal failure, the number
of patients with rebleeding was too small to allow adjustment for
other risk factors.7,22,23,41–44 Third, although we formulated precise
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 2, February 2015
diagnostic criteria for CDB, the number of patients who underwent
urgent colonoscopy was small (n = 6), which likely resulted in a low
detection rate of SRH. Nevertheless, the groups did not significantly
differ in urgent colonoscopy rate.
CONCLUSIONS
We found that high-dose barium therapy reduced recurrent
bleeding in a series of patients with CDB, particularly those with a
presumptive diagnosis and with spontaneous cessation. This treatment also decreased the frequency of rehospitalization, repeated
blood transfusion, and colonoscopic reexamination, and shortened
the length of hospitalization during long-term follow-up. Generalization of these results requires additional investigation at multiple
institutions with a larger group of patients.
ACKNOWLEDGMENT
We express our gratitude to Hisae Kawashiro (Clinical Research Coordinator). We thank all the staff of the endoscopy unit.
REFERENCES
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. Chait MM. Lower gastrointestinal bleeding in the elderly. World J Gastrointest
Endosc. 2010;2:147–154.
3. Strate LL, Ayanian JZ, Kotler G, et al. Risk factors for mortality in lower
intestinal bleeding. Clin Gastroenterol Hepatol. 2008;6:1004–1010; quiz 955.
4. McGuire HH, Jr, Haynes BW, Jr. Massive hemorrhage for diverticulosis of
the colon: guidelines for therapy based on bleeding patterns observed in fifty
cases. Ann Surg. 1972;175:847–855.
5. McGuire HH, Jr. Bleeding colonic diverticula. A reappraisal of natural history
and management. Ann Surg. 1994;220:653–656.
6. Poncet G, Heluwaert F, Voirin D, et al. Natural history of acute colonic diverticular bleeding: a prospective study in 133 consecutive patients. Aliment
Pharmacol Ther. 2010;32:466–471.
7. Niikura R, Nagata N, Yamada A, et al. Recurrence of colonic diverticular
bleeding and associated risk factors. Colorectal Dis. 2012;14:302–305.
8. Kaltenbach T, Watson R, Shah J, et al. Colonoscopy with clipping is useful
in the diagnosis and treatment of diverticular bleeding. Clin Gastroenterol
Hepatol. 2012;10:131–137.
9. Koperna T, Kisser M, Reiner G, et al. Diagnosis and treatment of bleeding
colonic diverticula. Hepatogastroenterology. 2001;48:702–705.
10. Bloomfeld RS, Rockey DC, Shetzline MA. Endoscopic therapy of acute diverticular hemorrhage. Am J Gastroenterol. 2001;96:2367–2372.
11. Ishii N, Setoyama T, Deshpande GA, et al. Endoscopic band ligation for colonic
diverticular hemorrhage. Gastrointest Endosc. 2012;75:382–387.
12. Fiorito JJ, Brandt LJ, Kozicky O, et al. The diagnostic yield of superior mesenteric angiography: correlation with the pattern of gastrointestinal bleeding. Am
J Gastroenterol. 1989;84:878–881.
13. Adams JT. Therapeutic barium enema for massive diverticular bleeding. Arch
Surg. 1970;101:457–460.
14. Chorost MI, Fruchter G, Kantor AM, et al. The therapeutic barium enema
revisited. Clin Radiol. 2001;56:856–858.
15. Fujimoto A, Sato S, Kurakata H, et al. Effectiveness of high-dose barium enema
filling for colonic diverticular bleeding. Colorectal Dis. 2011;13:896–898.
16. Iwamoto J, Mizokami Y, Shimokobe K, et al. Therapeutic barium enema for
bleeding colonic diverticula: four case series and review of the literature. World
J Gastroenterol. 2008;14:6413–6417.
17. Matsuhashi N, Akahane M, Nakajima A. Barium impaction therapy for refractory colonic diverticular bleeding. AJR Am J Roentgenol. 2003;180:490–492.
18. Pausawasdi N, Al-Hawary M, Higgins PD. Therapeutic high-density barium
enema in a case of presumed diverticular hemorrhage. Case Rep Gastroenterol.
2011;5:88–94.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
Barium Therapy for Diverticular Bleeding
19. Zuckerman GR, Prakash C. Acute lower intestinal bleeding. Part II: etiology,
therapy, and outcomes. Gastrointest Endosc. 1999;49:228–238.
20. Jensen DM, Machicado GA, Jutabha R, et al. Urgent colonoscopy for the
diagnosis and treatment of severe diverticular hemorrhage. N Engl J Med.
2000;342:78–82.
21. Lhewa DY, Strate LL. Pros and cons of colonoscopy in management
of acute lower gastrointestinal bleeding. World J Gastroenterol. 2012;18:
1185–1190.
22. Jansen A, Harenberg S, Grenda U, et al. Risk factors for colonic diverticular
bleeding: a Westernized community based hospital study. World J Gastroenterol. 2009;15:457–461.
23. Yamada A, Sugimoto T, Kondo S, et al. Assessment of the risk factors for
colonic diverticular hemorrhage. Dis Colon Rectum. 2008;51:116–120.
24. Williams SM, Harned RK. Recognition and prevention of barium enema complications. Curr Probl Diagn Radiol. 1991;20:123–151.
25. Tsuruoka N, Iwakiri R, Hara M, 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.
26. Nishikawa H, Maruo T, Tsumura T, et al. Risk factors associated with recurrent
hemorrhage after the initial improvement of colonic diverticular bleeding. Acta
Gastroenterol Belg. 2013;76:20–24.
27. Okamoto T, Watabe H, Yamada A, et al. The association between arteriosclerosis related diseases and diverticular bleeding. Int J Colorectal Dis.
2012;27:1161–1166.
28. Fujino Y, Inoue Y, Onodera M, et al. Risk factors for early re-bleeding and
associated hospitalisation in patients with colonic diverticular bleeding. Colorectal Dis. 2013;15:982–986.
29. Miura S, Kodaira S, Aoki H, et al. Bilateral type diverticular disease of the
colon. Int J Colorectal Dis. 1996;11:71–75.
30. Angtuaco TL, Reddy SK, Drapkin S, et al. The utility of urgent colonoscopy in
the evaluation of acute lower gastrointestinal tract bleeding: a 2-year experience
from a single center. Am J Gastroenterol. 2001;96:1782–1785.
31. Schmulewitz N, Fisher DA, Rockey DC. Early colonoscopy for acute lower GI
bleeding predicts shorter hospital stay: a retrospective study of experience in a
single center. Gastrointest Endosc. 2003;58:841–846.
32. Green BT, Rockey DC, Portwood G, et al. Urgent colonoscopy for evaluation
and management of acute lower gastrointestinal hemorrhage: a randomized
controlled trial. Am J Gastroenterol. 2005;100:2395–2402.
33. Strate LL, Syngal S. Timing of colonoscopy: impact on length of hospital stay in
patients with acute lower intestinal bleeding. Am J Gastroenterol. 2003;98:317–
322.
34. Noveroske RJ. Intracolonic pressures during barium enema examination. Am
J Roentgenol Radium Ther Nucl Med. 1964;91:852–863.
35. Browder W, Cerise EJ, Litwin MS. Impact of emergency angiography in massive lower gastrointestinal bleeding. Ann Surg. 1986;204:530–536.
36. Miller RE, Skucas J, Violante MR, et al. The effect of barium on blood in the
gastrointestinal tract. Radiology. 1975;117:527–530.
37. Gubler JA, Kukral AJ. Barium appendicitis. J Int Coll Surg. 1954;21:
379–384.
38. Maglinte DD, Bush ML, Aruta EV, et al. Retained barium n the appendix:
diagnostic and clinical significance. AJR Am J Roentgenol. 1981;137:529–
533.
39. Matrana MR, Margolin DA. Epidemiology and pathophysiology of diverticular
disease. Clin Colon Rectal Surg. 2009;22:141–146.
40. Loffeld RJ, Engel A, Dekkers PE. Incidence and causes of colonoscopic perforations: a single-center case series. Endoscopy. 2011;43:240–242.
41. Niikura R, Nagata N, Akiyama J, et al. Hypertension and concomitant arteriosclerotic diseases are risk factors for colonic diverticular bleeding: a casecontrol study. Int J Colorectal Dis. 2012;27:1137–1143.
42. Strate LL. Lifestyle factors and the course of diverticular disease. Dig Dis.
2012;30:35–45.
43. Strate LL, Liu YL, Aldoori WH, et al. Obesity increases the risks of diverticulitis and diverticular bleeding. Gastroenterology. 2009;136:115–122.e1.
44. Strate LL, Liu YL, Huang ES, et al. Use of aspirin or nonsteroidal antiinflammatory drugs increases risk for diverticulitis and diverticular bleeding.
Gastroenterology. 2011;140:1427–1433.
www.annalsofsurgery.com | 275
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.