Efficacy of Contrast-enhanced Computed Tomography for the Treatment Strategy of Colonic Diverticular Bleeding

Transcription

1 ORIGINAL ARTICLE Efficacy of Contrast-enhanced Computed Tomography for the Treatment Strategy of Colonic Diverticular Bleeding Tomoya Sugiyama, Yoshikazu Hirata, Yuki Kojima, Takuya Kanno, Mikitoshi Kimura, Yusuke Okuda, Kenichi Haneda, Hirokazu Ikeuchi, Tomohiro Morikawa, Hisato Mochizuki, Hiroki Takada and Satoshi Sobue Abstract Objective Diverticular bleeding is the most common cause of acute lower gastrointestinal bleeding, and its incidence has recently increased. However, the treatment strategy of diverticular bleeding has not yet been established. The aim of the study was to investigate the efficacy of contrast-enhanced computed tomography (CECT) to determine the indication for urgent colonoscopy to achieve hemostasis. Methods A total of 124 patients diagnosed with diverticular bleeding between 2012 and 2013 in our hospital were analyzed. The clinical behavior, factors related to detecting bleeding diverticula, and risk factors for early rebleeding of diverticular bleeding were evaluated. Results Clinical behavior: Bleeding diverticula were identified in 23 of 124 (19%) patients and most of them (16/23; 70%) were located in the ascending colon. Hemostasis was achieved in all 23 cases, however, six (26%) developed early rebleeding. Factors for detecting bleeding diverticula: In patients in whom extravasation was detected using CECT, the endoscopic detection rate of bleeding diverticula was 60% (12/20), while bleeding diverticula were detected in only 31% (11/35) of patients in whom extravasation was not detected using CECT (p<0.05). The interval between the first hematochezia and colonoscopy in which the bleeding point was detected by colonoscopy (median 23.5 hours) was shorter than that in which bleeding diverticula were not detected (median 43.6 hours) (p<0.01). Risk factors for short term rebleeding: Using a univariate analysis, atherosclerotic comorbidity, anti-inflammatory drugs including low-dose aspirin, antithrombotic agents, vital signs on admission, hemoglobin level on hospitalization, and extravasation on CECT were not found to be significant risk factors. Conclusion The finding of extravasation on CECT is the most important factor for identifying and treating bleeding diverticula by colonoscopy. In such cases, urgent colonoscopy is recommended. Key words: diverticular bleeding, contrast-enhanced computed tomography, extravasation, endoscopic hemostasis (Intern Med 54: , 2015) () Introduction Colonic diverticula are pseudodiverticula resulting from herniation of the mucosa and submucosa through a weakened portion of the colonic wall (1). Colonic diverticulosis is an acquired disease caused by an increased intestinal pressure due to the decreased intake of dietary fiber and mucosal fragility associated with aging (2). With changes in dietary habits and increasingly aging populations, there is a rising incidence of colonic diverticulosis in developed countries (2). In Western countries, diverticulosis typically affects the left side of the colon, whereas in Japan, it also affects the right side (2, 3). Although most patients with colonic diverticulosis are asymptomatic throughout their life, this disease occasionally Department of Gastroenterology, Kasugai Municipal Hospital, Japan Received for publication February 4, 2015; Accepted for publication May 21, 2015 Correspondence to Dr. Yoshikazu Hirata, yskzhtm4@gmail.com 2961

2 causes complications such as diverticulitis or bleeding. Diverticular bleeding has been reported to be the most common cause of lower gastrointestinal bleeding (4, 5) and accounts for >40% of lower gastrointestinal bleeding (5). Colonic diverticular bleeding typically has a mild course and stops spontaneously without therapeutic intervention in approximately 70% to 80% of cases (5-8). However, some cases require endoscopic, angiographic, and surgical treatment depending on the nature of the bleeding (5-8), and the diagnostic and therapeutic strategy for colonic diverticular bleeding has not yet been established. Recently, several studies have recommended urgent colonoscopy for the initial evaluation of diverticular bleeding, and endoscopic treatments including endoscopic clipping and endoscopic band ligation have been reported to be effective for hemostasis of diverticular bleeding (9-15). For the treatment of colonic diverticular bleeding, it is important to determine the indication for therapeutic intervention, especially urgent colonoscopy, and to detect bleeding diverticula. Obana et al. (16) reported that contrast-enhanced computed tomography (CECT) is useful for the detection of colonic diverticular bleeding. Considering these previous findings, we hypothesized that the strategy for treating diverticular bleeding may be established using extravasation detected on CECT. Furthermore, some patients experience short-term recurrent bleeding (during hospitalization), as well as long-term rebleeding. The rate of rebleeding after one episode has been reported to be between 13.8% and 38% (5, 10, 17, 18). Several reports identified the risk factors for rebleeding of colonic diverticular hemorrhage (19, 20), however, the factors related to short-term recurrent bleeding are still unclear. The purpose of this study was to evaluate the efficacy of CECT for detecting bleeding diverticula and to establish a strategy for the treatment of colonic diverticular bleeding. In addition, the risk factors for early rebleeding of diverticular bleeding were evaluated. Materials and Methods Study setting and patient population Data were obtained from the Department of Gastroenterology, Kasugai Municipal Hospital. The diagnosis of colonic diverticular bleeding was made according to the criteria previously reported by Jensen et al. (9). A patient with hematochezia was diagnosed as having colonic diverticular bleeding when active bleeding from a diverticulum was observed or when blood clots were found in the colon with diverticula, but no blood was in the terminal ileum, and no other demonstrable cause of bleeding was present. A total of 124 patients diagnosed with colonic diverticular bleeding were identified between April 2012 and December From April 2012 to December 2012, a large number of patients had not undergone CECT, and after January 2013, CECT was performed for most patients, excluding those with renal dysfunction. The risk factors for early rebleeding were also investigated. Early rebleeding was defined as diverticular rebleeding during hospitalization at least three days after the last hematochezia. CECT procedures A total of 90 ml of iopamidol (Iopamiron 300; Bayer HealthCare, Berlin, Germany) was power-injected intravenously at a rate of 1.5 ml/s. The patients were scanned using a 64-MDCT (Aquilion TM, Toshiba Medical Systems, Tochigi, Japan). Scanning was performed with the following parameters: 120 kv, 350 mas, 0.5 s per rotation, 0.5 mm collimation, and beam pitch. All patients underwent craniocaudal scanning in the supine position during a single breath hold. A single-phase scan was performed 90 s after the beginning of contrast material injection. Images were acquired with 0.5-mm section width and a reconstruction interval of 7 mm. Coronal multiplanar images were reconstructed at 0.5-mm section width and a 5-mm interval. Endoscopic procedures All patients underwent colonoscopy during hospitalization except for readmission for colonic diverticular bleeding within one year. When endoscopists were available at the time of admission, colonoscopy without any bowel preparation was attempted first. Otherwise, the patients were administered polyethylene glycol (PEG) solution before colonoscopy to prepare for the examination when possible. When colonoscopy revealed active bleeding from a diverticulum, endoscopic hemostasis with a hemoclip was performed directly on the vessel if possible. When direct clipping was not possible because of dome location, massive hemorrhage, or a small diverticular orifice, indirect clipping was selected, and bleeding diverticula were closed with multiple hemoclips in a zipper fashion (21). Monotherapy with clipping was performed, and injection of epinephrine solution was not performed. If rebleeding occurred, CECT was repeated, and endoscopic treatment was attempted first in cases of extravasation. Transarterial embolization (TAE) or colectomy was subsequently performed when diverticular bleeding was not controlled because of massive rebleeding from the previously treated diverticula, a poor endoscopic view, or hemodynamic instability. Outcomes The major outcomes of this study were the detection rate of bleeding diverticula using colonoscopy, contributing factors for detecting bleeding diverticula (extravasation of CECT, interval between hematochezia and CECT, or interval between bleeding and colonoscopy), and risk factors for early rebleeding (medication history, comorbidities, or vital signs and hemoglobin levels on admission). Statistical analysis Continuous variables are expressed as the mean values ± standard deviation (SD) and were compared between the 2962

3 Table 1. Patients Characteristics. All patients patient underwent ce-ct. n=55 extra(+) group extra(-) group n=124 n=20 n=35 p value a Mean age (±SD) year 72.5± ± ±8.8 Sex Male p=0.26 Female Comorbidity Hypertension p=0.56 Cerebrovascular disease p=0.58 Cardiovascular disease p=0.67 Diabetes mellitus p=0.68 Atrial fibrillation p=0.68 Chronic kidney disease p=0.26 Medication Aspirin p=0.63 Anticoagulants NSAIDs p=0.06 the number of patient underwent urgent colonoscopy p=0.33 Past history of diverticular bleeding p<0.05 SD : standard deviation, NSAIDs : Non-steroidal anti-inflammatory drugs ce-ct : Contrast enhanced-computed tomography extra : extravasation a p value reflects comparison between extravasation(+) and extravasation(-) groups. two groups using the Mann-Whitney U test and Student s t- test. Categorical variables were compared among the groups using the chi-square test. Yates correction was applied when appropriate. All statistics were calculated using the R software program, version 2.92 (R Development Core Team, 2009). Probabilities of less than 0.05 were considered to be statistically significant. Patients characteristics Results From April 2012 to December 2013, 124 patients were admitted to our hospital because of colonic diverticular bleeding, and all of these patients were enrolled in this study. The clinical features of these patients are presented in Table 1. In this group, the median age was 72.5±10.8 years. Many of the patients had some atherosclerotic comorbidities, such as hypertension (n=41, 33%), cerebrovascular disease (n=23, 18.5%), and cardiovascular disease (n=24, 19.5%). Low-dose aspirin was prescribed for 34 patients (28%), anticoagulant drugs were prescribed for seven patients (6%), and non-steroidal anti-inflammatory drugs (NSAIDs) were prescribed for 12 patients (10%). Fifty-eight patients (46.7%) had a previous history of colonic diverticular bleeding. CECT was carried out for 55 patients. The clinical characteristics of the extra(+) group (e.g., cases where CECT showed extravasation) and the extra(-) group (e.g., cases where CECT did not show extravasation) were compared. There were no significant difference in the baseline characteristics between the extra(+) and extra(-) groups (Table 1). Clinical behavior Ninety-nine patients underwent colonoscopy, and bleeding diverticula were identified in 23 patients (23%). Bleeding diverticula were located in the ascending colon in 16 patients (70%), the sigmoid colon in four patients (17%), the descending colon in two patients (9%), and the transverse colon in one patient (4%) (Table 2). All 23 patients underwent endoscopic hemostatic clipping, however, six patients (26%) developed rebleeding, while 12 patients (16%) who did not need endoscopic hemostasis developed short-term rebleeding (Table 2). The ratio (number) of patients with bleeding diverticula with or without PEG ingestion was also investigated. There was no significant difference in the bleeding diverticulum detection rate between the with PEG and without PEG groups (Table 2). Surgical treatment was required for five patients. In all cases, bleeding diverticula were detected using CECT in the ascending colon and endoscopic hemostasis was performed. In case 1, the patient was strongly anxious about potential rebleeding following the endoscopic procedure and surgical treatment was performed according to his wishes. Despite the conservative treatment, including a blood transfusion after the endoscopic treatment, diverticular bleeding was not controlled and surgical treatment was performed in the other cases. TAE was carried out in one case (case 2). Unfortunately, a bleeding artery was not detected during the TAE procedure, thus embolization could not performed because of the risk of ischemia (Table 3). 2963

4 Table 2. Clinical Behavior of 99 Patients who Underwent Colonoscopy. bleeding diverticula detection rate (n=99) 23% (23/99) p value Location of bleeding diverticula (n=23) ascending colon 70% (n=16) sigmoid colon 17% (n=4) descending colon 9% (n=2) transvers colon 4% (n=1) The ratio (number) of short term rebleeding patients undergone colonoscopy (n=99) The ratio (number) of patient detected bleeding diverticula with or without PEG (n=99) PEG: polyethylene glycol endoscopic hemostasis(+) group 26% (6/23) 0.38 endoscopic hemostasis(-) group 16% (12/76) with PEG 21% (16/75) 0.23 without PEG 33% (8/24) Table 3. The Patient who Required Surgical Treatment. case 1 case2 case3 case4 case5 Age Gender M F M M F blood transfusion(u) extravasation of ce-ct location of bleeding diverticulum A-colon A-colon A-colon A-colon A-colon endoscopic detection of bleeding point endoscopic hemstasis success Failure Failure Failure Failure TAE not performed Failure not performed not performed not performed operation ileocecal resection ileocecal resection right hemicoloctomy ileocecal resection right hemicoloctomy ce-ct: Contrast enhanced-computed tomography TAE: transarterial embolization A-colon: ascending colon Factors for detecting bleeding diverticula The detection rates of bleeding diverticula by colonoscopy in the extra(+) group and the extra(-) group were compared. In the extra(+) group, bleeding diverticula were detected in 60% (12/20) of patients, whereas bleeding diverticula were detected in only 31% (11/35) of the patients in the extra(-) group (p<0.05) (Table 4). The concordance rate of the CT findings and the colonoscopic findings was 95.7% (22/23). The median interval time from the appearance of the first hematochezia to CECT examination was compared between the extra(+) group and the extra(-) group. In the extra(+) group, the median interval time was 4.8±4.71 hours, while in the extra(-) group, the median interval time was 7.5±6.9 hours. The correlation between the interval time to CECT and the detection rate of extravasation was not significant. The correlation between the interval time from the first hematochezia to urgent colonoscopy and the detection of bleeding diverticula was investigated. In the group in which bleeding diverticula were detected, the median interval time was 23.5±25.7 hours. On the other hand, in the group in which bleeding diverticula were not detected, the median interval time was 43.6±32.5 hours, and the difference in the interval from the first hematochezia to colonoscopy was significant (p<0.01). Risk factors for early rebleeding A univariate analysis showed that the body mass index (BMI), past comorbidity of hypertension and chronic kidney disease, NSAID use, low-dose aspirin use, anticoagulant therapy, systolic blood pressure and heart rate on admission, hemoglobin level on hospitalization, and extravasation of CECT were not factors contributing to early rebleeding (Table 5). Discussion The present study indicated that: 1) most bleeding diverticula are located in the ascending colon; 2) patients who showed extravasation on CECT were highly likely to show bleeding diverticula on colonoscopy; 3) urgent colonoscopy should be performed as soon as possible; and 4) no significant risk factors for short-term rebleeding were identified. In Western countries, diverticula are predominantly observed in the left side of the colon, while right-sided diverticula are more commonly found in Asian countries (22). In the present study, approximately 70% of bleeding diverticula were present in the right-side of the colon. These data are in agreement with the previous report that 71% of bleeding points were located in the right colon in colonic diverticular bleeding (21). Our findings did not demonstrate the efficacy of PEG ingestion for detecting bleeding diverticula, although Obana et al. (16) previously reported that urgent colonoscopy should be performed after bowel preparation using PEG ingestion to detect bleeding diverticula. However, the efficacy of bowel preparation remains controversial, thus 2964

5 Table 4. Correlation between Extravasation on Ce-CT and Detecting the Bleeding Diverticulum on Colonoscopy. Bleeding diverticulum detected Bleeding diverticulum not detected Extra (+) group Extra (-) group *p< # The concordance rate of the CT and colonoscopic findings is 95.7% (22/23) *Chi-square for independence test. ce-ct : Contrast enhanced-computed tomography, Extra : extravasation Table 5. Risk Factors for Short Term Re-bleeding. Factor Re-bleeding(+) Re-bleeding(-) *p value Body mass index 25 kg/m 2 No Yes Hypertension No Yes Chronic Kidney Disease No Yes NSAIDs No Yes Low-dose aspirin No Yes Anticoagulant therapy No Yes BP 80 mmhg No Yes HR 100/min No 3 12 Yes Hb 10 g/dl No Yes Extravasation on ce-ct No 4 31 Yes *Chi-square for independence test. NSAIDs : Non-steroidal anti-inflammatory drugs, BP : Blood pressure, HR : heart rate, ce-ct : Contrast enhanced-computed tomography further examination is needed. In the treatment of colonic diverticular bleeding, it is most important to detect bleeding diverticula. The information on the bleeding source is crucial for making therapeutic decisions such as colonoscopy, TAE and operation (16). According to the previous reports, colonoscopic clipping may be the most effective and safest treatment (11, 13, 23). Selective angiography techniques using smaller catheters to embolize the bleeding artery have shown high segmental ischemia in up to 20% of patients (24). In addition, there are a fewer number of radiologists, compared with endoscopists, who can accurately treat colonic diverticular bleeding using the TAE technique. The operative mortality with surgery can be very significant, up to 10% when the bleeding is localized and up to 75% when it is not and subtotal colectomy is performed (25). Taking into consideration these facts, endoscopic hemostasis should be utilized first in the treatment of colonic diverticular bleeding (11-13). However, considering the fact that spontaneous resolution of colonic diverticular bleeding occurs in 70-80% of cases without therapeutic intervention (6), the patient group that requires endoscopic intervention should be identified. The present data showed that patients who showed extravasation of the contrast medium on a CT examination were significantly more likely to show bleeding diverticula on colonoscopy (60%) than those who were negative on CECT (31%). In addition, the concordance rate of the CT and colonoscopic findings was 95.7%. According to these data, urgent colonoscopy is recommended for patients who show extravasation on CECT. In the present study, there was a significant correlation between the interval from the first hematochezia to colonoscopy and the detection rate of bleeding diverticula. Considering these data, it is important that urgent colonoscopy should be performed as soon as possible for patients who show extravasation of the contrast medium on CECT in cases of colonic diverticular bleeding. According to our findings, the strategy for colonic diverticular bleeding in our hospital is presented in Figure. The most important initial step is to understand the patient s general status and stabilize the vital signs with anti-shock management. CECT should be subsequently performed to detect bleeding diverticula, excluding patients in whom it is contraindicated, and then decided whether urgent colonoscopy is required. If extravasation is seen, then urgent colonoscopy should be performed as soon as possible. Endoscopic hemostasis using clipping techniques would be performed first in the treatment of colonic diverticular bleeding. If the hematochezia continues, then emergent TAE or colectomy should be performed (Figure). If no extravasation is detected on CECT, then conservative treatment is recommended. The present data demonstrated that short-term rebleeding occurred in 18.5% of cases, and the rate was relatively high 2965

6 Div r st vit si A - ther py Extravasation (+) Ur e t opy Det poi t Hemost o tr -CT TAE r tre tme t St vit si Extravasation (-) rv ve tre tme t Re i Figure. The treatment strategy for colonic diverticular bleeding. CECT should be performed at first. If extravasation is seen, urgent colonoscopy should be performed. (26%) in the group that underwent colonoscopic hemostasis (p=0.08). These data suggest that it is important to identify the risk factors for rebleeding when managing diverticular bleeding patients, especially in those for whom endoscopic hemostasis is necessary. NSAIDs use, antiplatelet drugs, hypertension, atherosclerotic diseases, and serum creatinine level were mentioned as risk factors (19, 20, 26). However, most of these previous studies referred to long-term rebleeding, and therefore the risk factors for short-term rebleeding are still unclear. Fujino et al. (27) reported that the evidence of shock on admission was the only risk factor. In the present study, no significant risk factors for early rebleeding were identified, including comorbidity, NSAIDs use, lowdose aspirin, vital signs on admission, laboratory findings, andextravasationoncect. There are several limitations associated with this study. First, it was a retrospective, cohort study involving colon diverticular bleeding patients in a single hospital. Second, the number of cases might have been too small to identify the predictive factors of CECT findings and to investigate the risk factors for early rebleeding. Third, most patients were hemodynamically stable at the time of CECT and colonoscopy was performed. In conclusion, the finding of extravasation on CECT is the most important factor to identify and treat bleeding diverticula using colonoscopy. In such cases, urgent colonoscopy is recommended. The authors state that they have no Conflict of Interest (COI). References 1. Meyers MA, Alonso DR, Baer JW. Pathogenesis of massively bleeding colonic diverticulosis: new observations. AJR Am J Roentgenol 127: , Miura S, Kodaira S, Shatari T, Nishioka M, Hosoda Y, Hisa TK. Recent trends in diverticulosis of the right colon in Japan: retrospective review in a regional hospital. Dis Colon Rectum 43: , Stollman N, Raskin JB. Diverticular disease of the colon. Lancet 363: , Ohi G, Minowa K, Oyama T, et al. Changes in dietary fiber intake among Japanese in the 20th century: a relationship to the prevalence of diverticular disease. Am J Clin Nutr 38: , Longstreth GF. Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: a populationbased study. Am J Gastroenterol 92: , McGuire HH Jr. Bleeding colonic diverticula. A reappraisal of natural history and management. Ann Surg 220: , Funaki B, Kostelic JK, Lorenz J, et al. Superselective microcoil embolization of colonic hemorrhage. AJR Am J Roentgenol 177: , Tan KK, Nallathamby V, Wong D, Sim R. Can superselective embolization be definitive for colonic diverticular hemorrhage? An institution s experience over 9 years. J Gastrointest Surg 14: , Jensen DM, Machicado GA, Jutabha R, Kovacs TO. Urgent colonoscopy for the diagnosis and treatment of severe diverticular hemorrhage. N Engl J Med 342: 78-82, Bloomfeld RS, Rockey DC, Shetzline MA. Endoscopic therapy of acute diverticular hemorrhage. Am J Gastroenterol 96: , Simpson PW, Nguyen MH, Lim JK, Soetikno RM. Use of endoclips in the treatment of massive colonic diverticular bleeding. Gastrointest Endosc 59: , Yen EF, Ladabaum U, Muthusamy VR, Cello JP, McQuaid KR, Shah JN. Colonoscopic treatment of acute diverticular hemorrhage using endoclips. Dig Dis Sci 53: , Kaltenbach T, Watson R, Shah J, et al. Colonoscopy with clipping is useful in the diagnosis and treatment of diverticular bleeding. Clin Gastroenterol Hepatol 10: , Setoyama T, Ishii N, Fujita Y. Enodoscopic band ligation (EBL) is superior to endoscopic clipping for the treatment of colonic diverticular hemorrhage. Surg Endosc 25: , Ishii N, Setoyama T, Deshpande GA, et al. Endoscopic band ligation for colonic diverticular hemorrhage. Gastrointest Endosc 75: , Obana T, Fujita N, Sugita R, et al. Prospective evaluation of contrast-enhanced computed tomography for the detection of colonic diverticular bleeding. Dig Dis Sci 58: , Czymek R, Kempf A, Roblick U, et al. Factors predicting the postoperative outcome of lower gastrointestinal hemorrhage. Int J Colorectal Dis 24: , Poncet G, Heluwaert F, Voirin D, Bonaz B, Faucheron JL. Natural history of acute colonic diverticular bleeding: a prospective study in 133 consecutive patients. Aliment Pharmacol Ther 32: , Nishikawa H, Maruo T, Tsumura T, Sekikawa A, Kanesaka T, Osaki Y. Risk factors associated with recurrent hemorrhage after the initial improvement of colonic diverticular bleeding. Acta Gastroenterol Belg 76: 20-24, Okamoto T, Watabe H, Yamada A, Hirata Y, Yoshida H, Koike K. The association between arteriosclerosis related diseases and diverticular bleeding. Int J Colorectal Dis 27: , Ishii N, Hirata N, Omata F, et al. Location in the ascending colon is a predictor of refractory colonic diverticular hemorrhage after endoscopic clipping. Gastrointest Endosc 76: , Kang JY, Melville D, Maxwell JD. Epidemiology and management of diverticular disease of the colon. Drugs Aging 21: , Hokama A, Uehara T, Nakayoshi T, et al. Utility of endoscopic hemoclipping for colonic diverticular bleeding. Am J Gastroenterol 92: , Khanna A, Ognibene SJ, Koniaris LG. Embolization as first-line therapy for diverticulosis-related massive lower gastrointestinal bleeding: evidence from a meta-analysis. J Gastrointest Surg 9: , Chen CY, Wu CC, Jao SW, Pai L, Hsiao CW. Colonic diverticular 2966

7 bleeding with comorbid diseases may need elective colectomy. J Gastrointest Surg 13: , Niikura R, Nagata N, Yamada A, Akiyama J, Shimbo T, Uemura N. Recurrence of colonic diverticular bleeding and associated risk factors. Colorectal Dis 14: , Fujino Y, Inoue Y, Onodera M, et al. Risk factors for early rebleeding and associated hospitalization in patients with colonic diverticular bleeding. Colorectal Dis 15: , The Japanese Society of Internal Medicine