ORIGINAL PAPERS

The association between a positive lactulose methane breath test and in constipated patients

Ji-Min Lee, Chang-Nyol Paik, Yeon-Ji Kim, Dae-Bum Kim, Woo-Chul Chung, Kang-Moon Lee and Jin-Mo Yang Department of Internal Medicine. College of Medicine. St. Vincent’s Hospital. The Catholic University of Korea. Suwon, Korea

Received: 18/04/2017 · Accepted: 18/10/2017 Correspondence: Chang-Nyol Paik. Division of . Department of Internal Medicine. St. Vincent’s Hospital. College of Medicine. The Catholic University of Korea. 93 Jungbu Daero (Ji-dong), Suwon Si, Paldal-gu. 16247 Gyeonggi-Do, South Korea. e-mail: [email protected]

ABSTRACT males. Most rectoceles are small and asymptomatic, while some rectoceles, especially large ones, are associated with Objectives: Rectocele with might be related a difficult evacuation or with chronic constipation (1). How- ever, the etiology and physiology of rectoceles are unclear to methane (CH4) producing intestinal bacteria. We investi- and it is still not fully understood whether rectocele is a gated the breath CH4 levels and the clinical characteristics of colorectal motility in constipated patients with rectocele. cause or a result of constipation.

Methods: A database of consecutive female outpatients Functional constipation is potentially associated with intes-

was reviewed for the evaluation of constipation according tinal methane (CH4) according to breath testing and with to the Rome III criteria. The patients underwent the lactulose delayed intestinal transit. Methane produced by enteric bac- teria can slow small bowel transit and increase intestinal CH4 breath test (LMBT), colon marker study, anorectal ma- nometry, and bowel symptom questionnaire. contractility (2-4). A positive lactulose breath test (LBT) for

The profiles of the lactulose breath test (LBT) in 33 patients methane (CH4) (LMBT+), which indicates the presence of with rectocele (with size ≥ 2 cm) and 26 patients with func- small intestinal bacterial overgrowth (SIBO) that produces

tional constipation (FC) were compared with the breath test CH4, has been recently related to functional constipation results of 30 healthy control subjects. with delayed colonic transit (3-5). We hypothesize that a

relationship could exist between breath CH4 levels and rec- Results: The mean size of rectocele was 3.52 ± 1.06 cm. The tocele in constipated female patients. rate of a positive LMBT (LMBT+) was significantly high- er in patients with rectocele (33.3%) than in those with FC The aims of the study were to compare the profiles of LMBT in constipated patients with rectocele to patients with func- (23.1%) or healthy controls (6.7%) (p = 0.04). Breath CH4 concentration was positively correlated with rectosigmoid tional constipation or healthy controls. In addition, the char- colon transit time in rectocele patients (γ = 0.481, p < 0.01). acteristics of clinical symptoms, colonic transit time and A maximum high pressure zone pressure > 155 mmHg was anorectal pressure according to LMBT+ in constipated pa- a significant independent factor of LMBT+ in rectocele pa- tients with rectocele were evaluated. tients (OR = 8.93, 95% CI = 1.14-71.4, p = 0.04).

Conclusions: LMBT+ might be expected in constipated pa- MATERIALS AND METHODS tients with rectocele. Moreover, increased rectosigmoid co- lonic transit or high anorectal pressure might be associated This study was performed by a retrospective review of

with CH4 breath levels. Breath CH4 could be an important medical charts. The study was approved by the Institutional therapeutic target for managing constipated patients with Research Ethics Board of the Catholic University of Korea rectocele. (VC16RISI0158) and adheres to the guidelines of the Decla- ration of Helsinki. Key words: Breath methane. Constipation. Rectocele. Lac- tulose breath test.

Lee JM, Paik CN, Kim YJ, Kim DB, Chung WC, Lee KM, Yang JM. The as- INTRODUCTION sociation between a positive lactulose methane breath test and rectocele in constipated patients. Rev Esp Enferm Dig 2018;110(2):115-122. An anterior rectocele is an abnormal protrusion of the rectal wall, typically toward the posterior vagina, which occurs DOI: 10.17235/reed.2017.5017/2017 during . The condition frequently occurs in fe-

REV ESP ENFERM DIG 2018:110(2):115-122 1130-0108/2018/110/2/115-122 • REVISTA ESPAÑOLA DE ENFERMEDADES DIGESTIVAS DOI: 10.17235/reed.2017.5017/2017 © Copyright 2018. SEPD y © ARÁN EDICIONES, S.L. 116 J.-M. Lee et al.

Study population greater than 10 ppm or an increase in CH4 concentration of more than 10 ppm above baseline within 90 min (LMBT+ A prospective database of consecutive outpatients who vis- group). The LBT (mixed) + group was defined as satisfying ited the teaching hospital of the Catholic University School both criteria. of Medicine at St. Vincent’s Hospital for the evaluation of functional constipation between February 2012 and April 2016 was retrospectively reviewed. Female patients with Assessment of abdominal symptoms functional constipation over 18 years of age who underwent

the hydrogen (H2) - CH4 LBT, colon marker study, anorec- Demographic data were collected and a bowel symptom tal manometry, and defecography were enrolled into the questionnaire was routinely performed during LBT in or- study. Patients with functional constipation were identified der to diagnose patients with clinically suspected functional according to the Rome III criteria during the preceding three constipation. The Korean version of the Irritable Bowel Syn- months, with symptom at onset at least six months before drome Quality of Life questionnaire, which consists of the diagnosis (5-7). At least two of the following symptoms Rome III criteria together with additional questions about were required for the diagnosis of functional constipation: bowel symptoms, was used. This questionnaire has been a) straining during ≥ 25% of defecation; b) lumpy or hard used and validated in other studies (9-11). In addition, 13 stools in ≥ 25% of defecation; c) sensation of incomplete questions with regard to various bowel symptoms expe- evacuation in ≥ 25% of defecation; d) sensation of anorectal rienced during the preceding four weeks were included. obstruction/blockage in ≥ 25% of defecation; e) need for Information about symptoms of abdominal discomfort, manual maneuvers to facilitate defecation in ≥ 25% of cas- pain or cramps, hard or lumpy stools, lose or watery stools, es; and/or f) fewer than three per week. Among straining during a bowel movement, having to rush to the constipated patients, subjects with rectocele were included toilet for a bowel movement (urgency), passing mucus if the size of the rectocele was ≥ 2 cm, as diagnosed by during a bowel movement, abdominal fullness/ or defecography (8). swelling, passing gas (flatus), or chest pain, feel- ing full soon after starting a meal, passing urine frequently Patients were excluded if they had a history of diabetes and was collected. The severity of symptoms was mellitus, connective tissue disease, thyroid disease or gas- evaluated by the total symptom score, which was defined trointestinal surgery. Other exclusion criteria included the as the sum of the scores for symptom frequency and intru- use of anti-secretory agents such as proton pump inhibitors siveness. The frequency and intrusiveness of each symp-

or histamine (H2) receptor antagonists, antibiotics, probi- tom were assessed by each patient using a seven-point otics, prokinetics, narcotics, , bulking agents or scale from 0 (never) to 6 (always or extremely). As the total antidiarrheal drugs. Gastrointestinal disease, renal insuf- symptom score was defined as the sum of the scores for ficiency, disease, major psychiatric disease, hearing symptom frequency and intrusiveness, the range of total impairment masticatory dysfunction, colonoscopy within symptom scores for each symptom was 0-12. the previous three months or incomplete data were also exclusion criteria. The LBT profiles of female patients with constipated rectocele or functional constipation were com- Anorectal manometry pared with 30 (historic) healthy female controls. The control subjects were enrolled in order to determine normal LBT The patients underwent anal manometry using a wa- values using data from the Catholic University of Medicine ter-perfused catheter with eight radially aligned channels from 2007 (4). attached to a hydraulic capillary infusion system (Medtronic Inc., Minneapolis, MN, USA). The catheter was 4.5 mm in diameter with side holes of 0.8 mm in diameter. The ex- Lactulose breath tests amination was performed in the left lateral position using the rapid pull-through technique. The variables of basal and Lactulose breath tests were performed with gas chroma- maximum resting pressure, anal canal length, resting and tography equipment (Quintron BreathTracker SC; Quintron maximum high pressure zone (HPZ) pressure, HPZ length, Instrument Company, Milwaukee, WI, USA) following an basal and maximum squeezing pressure and sensory vol- overnight fast of at least 12 hours. Subjects were asked to ume were recorded and analyzed using data collected in the follow a strict low residual diet the day before the examina- Polygram software (version 4.1; Medtronic Inc.). tion. Thirty minutes prior to the breath test, patients were instructed to wash their mouths with 20 ml of 0.05% chlor- The manometry probe was inserted into the rectum and hexidine. Physical exercise and cigarette smoking were not orientated so that the most distal sensor (1 cm level) was permitted for 30 minutes prior to and during the test. After located posterior, at 1 cm from the anal verge. After probe collecting a baseline breath sample, the patients ingested placement, there was a rest period of about five minutes 10 g of lactulose syrup (lactulose concentrate; JW Phar- in order to give the patient time to relax and the sphinc- maceutical, Seoul, Korea). Duplicate samples of end expi- ter tone to return to basal levels. Resting pressure was ratory breath air were collected at baseline and at 15-min defined as the difference between the intrarectal pressure

intervals for 180 min. The breath H2 and CH4 concentrations and the maximum anal sphincter pressure at rest (Fig. 3). were evaluated with the equipment. The definition of an After probe placement, the maximum sphincter pressure LBT positive status indicating SIBO was defined and clas- was measured as an average of a one minute segment

sified as follows (9): a) a baseline 2H concentration greater at each level, i.e. at 1, 2 and 3 cm from the anal verge.

than 20 ppm or an increase in H2 concentration of more than The mean of the three highest values observed at any

20 ppm above baseline within 90 min (lactulose H2 breath site in the anal canal was taken as the maximum resting

test (LHBT) + group); or b) a baseline CH4 concentration pressure.

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017 The association between a positive lactulose methane breath test and rectocele in constipated patients 117

The patient was asked to squeeze the anus for as long as with the demographic data and LMBT profiles of healthy con- possible, for a maximum of 30 seconds, followed by a one trols. Clinical evaluations included age, body mass index (BMI), minute rest. The maximum squeeze pressure was defined intestinal symptoms, parameters of anorectal manometry and

as the difference between the intrarectal pressure and the colon transit time, type of constipation and breath H2 or CH4 highest pressure recorded at any level within the anal ca- levels according to the presence of rectocele or to a LMBT+ re- nal during the squeezing maneuver. By scanning the two sult. Continuous data were expressed as mean ± SD and were squeeze attempts, the mean of the highest pressures re- analyzed using independent-sample t-tests, whereas the cat- corded at any site in the anal canal was used to calculate egorical variables were expressed as a quantity (i.e., analyzed the maximum squeeze pressure (12). using χ2 tests or Fisher’s exact tests). The Pearson coefficient (γ) was used for the evaluation of the correlation between to-

tal breath CH4 concentration and colonic transit time. Multiple Colon maker study stepwise logistic regression analysis was used to identify the independent factors associated with LMBT+. A p value less Twenty radio-opaque markers within one capsule (Kolon mark- than 0.05 was considered to be significant for all tests. er; M.I. Tech, Pyeongtaek, Korea) were used to measure bowel transit time. Subjects swallowed one capsule per day in the morning for three consecutive days. Two plain abdominal radio- RESULTS graphs were taken on days 4 and 7 (13). The location of markers was classified by bony structures on the abdominal films. The Study populations markers located to the right of the vertebral spinous processes and above a line from the fifth vertebrae to the pelvic outlet During the study period, 39 rectocele patients and 44 pa- were included in the right colonic region. The markers on the tients with functional constipation were enrolled in the left of the vertebral spinous processes and above an imaginary study. Among the patients with rectocele, six were exclud- line from the fifth lumbar vertebrae to the anterior superior iliac ed from the analysis as they were male (n = 4) or due to a crest were included in the left colonic region. The markers that lack of data from day 7 of the colon marker study (n = 2). In were lower than a line from the pelvic brim on the right side contrast, among the patients with functional constipation, and the superior iliac crest on the left side were assigned to 18 were excluded as they were male (n = 14) or due to a lack the rectosigmoid region. Markers were counted in each of the of data from day 7 of the colon marker study (n = 4). Thir- right, left and rectosigmoid regions. The mean segmental and ty-three rectocele patients and 26 patients with functional total colonic transit times were calculated as 1.2 xⅹthe sum of constipation were finally analyzed. The age of patients with markers in the segmental colon on the radiographs on days 4 rectocele or functional constipation was significantly higher and 7, according to a formula described previously (14-16). Sub- than the age of the healthy controls (Table 1). The mean size jects were asked to perform their daily routines, to eat normal of rectoceles in cases was 3.52 ± 1.06 cm. diets, and to avoid excessive fiber intake and unusually inten- sive physical activity. Types of constipation were defined as: a) normal transit constipation; and b) delayed transit constipation. Comparison of lactulose breath test results in patients with The retention of at least 20% of the markers (n = 12) on day 7 rectocele, functional constipation and healthy controls was defined as delayed transit constipation (13). No differences were observed between patients with rectocele

and control subjects in the breath H2 concentration at any time Defecography points, except at 30 and 45 min (Fig. 1). With regard to breath

CH4, there were significant differences or different tendencies Patients fasted from the evening before the procedure and between patients with rectocele and control subjects at all of performed a rectal cleaning at home a few hours be- the time points (Fig. 2). Moreover, different tendencies were fore going to the hospital. At the beginning of the examina- demonstrated at time point 0, 30, 75, 90, and 105 min between tion, the patient was positioned on the left side and around patients with rectocele and functional constipation (Fig. 2). The 300 ml of thick barium paste was injected into the rectum rate of positivity of the LBT was 48.5% (16/33), 34.6% (9/26) using a plastic syringe connected to a catheter. In female and 26.7% (8/30) in subjects with rectocele, functional con- patients, the vagina was opacified with a commercially avail- stipation and healthy controls, respectively. The LMBT+ and able barium sulfate paste for oral use. The fluoroscopic table total CH4 concentration were significantly higher in patients was tilted vertically and a special commode was attached with rectocele than in patients with functional constipation or to the footboard with two or three water-filled annular pil- in the healthy controls (Table 1). The total CH4 concentration lows. The patient was then asked to sit on the commode at a was significantly higher in patients with rectocele than in the right lateral projection. The patient was instructed to empty healthy controls and the LMBT+ was significantly higher than the rectum completely and without interruption; this pro- in patients with functional constipation or healthy controls (Ta- cess took less than 30 seconds in physiological conditions. ble 1). In contrast, no differences were observed with regard An outpouching of the anterior rectal wall that bulged and to the rates of positivity of the LBT, LHBT or the total breath dislocated the opacified vaginal lumen during straining and H2 concentration among the groups. evacuation was observed in patients with rectocele (8).

Characteristics in constipated patients including rectocele Analysis and functional constipation

Demographic data and the LMBT profiles of patients with con- There were no differences between the patients with recto- stipated rectocele or functional constipation were compared cele and functional constipation with regard to total symptom

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017 118 J.-M. Lee et al.

Table 1. Demographic clinical data of controls, patients with functional constipation and rectocele

Rectocele Functional Controls p-value (n = 33) constipation (n = 26) (n = 30) Age (year) 57.68 ± 12.64 51.85 ± 14.43 40.33 ± 14.98 < 0.01 T2 A A B BMI 21.77 ± 1.94 22.18 ± 2.34 21.77 ± 2.74 0.76

Total H2, ppm 291.72 ± 247.45 263.00 ± 206.91 228.15 ± 193.35 0.52

Total CH4, ppm 166.21 ± 248.93 96.03 ± 94.43 40.63 ± 136.38 0.02 T2 A A,B B Positive LBT (%) 16 (48.5) 9 (34.6) 8 (26.7) 0.19

H2 5 (15.2) 3 (11.5) 6 (20.0) 0.24

CH4 6 (18.2) 2 (7.7) 1 (3.3) Mixed 5 (15.2) 4 (15.4) 1 (3.3) Positive LHBT (%) 10 (30.3) 7 (26.9) 7 (23.3) 0.82 Positive LMBT (%) 11 (33.3) 6 (23.1) 2 (6.7) 0.04 Total symptom scores 58.29 ± 28.73 61.26 ± 32.63 0.75 Frequency score 31.38 ± 14.51 32.05 ± 15.71 0.88 Bothersome score 26.92 ± 14.81 29.21 ± 17.65 0.65 Anorectal manometry3 Basal resting pressure 73.34 ± 26.42 73.45 ± 31.51 0.16 Maximum resting pressure 160.00 ± 56.86 166.14 ± 64.18 0.30 Anal canal length, cm 3.89 ± 1.06 3.64 ± 1.01 0.59 Resting HPZ pressure 102.47 ± 32.41 99.32 ± 38.94 0.13 Maximum HPZ pressure 143.17 ± 43.10 155.58 ± 55.62 0.11 Basal squeezing pressure 104.44 ± 29.87 113.37 ± 42.11 0.35 Maximum squeezing pressure 221.03 ± 56.23 251.18 ± 89.74 0.14 HPZ length, cm 1.93 ± 0.47 1.82 ± 0.57 0.73 Minimum sensory volume 58.79 ± 34.07 67.69 ± 44.30 0.76 Urgency volume 97.27 ± 40.64 99.62 ± 50.08 0.68 Maximum tolerable volume 153.94 ± 60.46 153.46 ± 62.67 0.65 Transit time on day 7 (hrs.) Right colon 15.16 ± 14.25 17.40 ± 22.27 0.64 Left colon 23.20 ± 18.23 19.43 ± 16.72 0.42 Rectosigmoid colon 18.62 ± 18.41 18.28 ± 17.43 0.94 Total colon time 56.98 ± 40.39 55.11 ± 42.31 0.86 Types of constipation, n (%) Normal transit (n = 41) 24 (72.7) 17 (65.4) 0.54 Delayed transit (n = 18) 9 (27.3) 9 (34.6) Data are expressed as mean ± SD or number (%). 1Statistical significance was tested by one-way analysis of variances or the chi-square test among groups.2 The same letters indicate non-significant 3 2 differences between groups based on Tukey’s multiple comparison test. Pressure (mmHg), length (cm) and volume (ml). BMI: body mass index (kg/m ); H2: hydrogen; CH4: methane; LBT: lactulose breath test; LHBT: lactulose hydrogen breath test; LMBT: lactulose methane breath test; HPZ: high pressure zone.

scores, anorectal manometry parameters, colon transit time was a positive correlation between total CH4 concentration and type of constipation (Table 1). With regard to a positive and left colonic transit time (γ = 0.27, p = 0.04), rectosigmoid LMBT, no differences were observed in age, BMI, presence of colonic transit time (γ = 0.30, p = 0.02) and total colonic transit rectocele, total symptom score, anorectal manometry param- time (γ = 0.31, p = 0.02). Delayed transit constipation was the eters and colon transit time. However, differences were ob- only significant independent factor of LMBT+ according to served with regard to the type of constipation (Table 2). There multivariable logistic regression analysis (Table 3).

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017 The association between a positive lactulose methane breath test and rectocele in constipated patients 119

Fig. 1. Lactulose hydrogen (H2) profi les (ppm, parts per Fig. 2. Lactulose methane (CH4) profi les (ppm, parts per million) million) in constipated patients with rectocele, patients with in constipated patients with rectocele, patients with functional functional constipation and control subjects (*p < 0.05). constipation and control subjects (* or $, p < 0.1; **p < 0.05). Table 2. Characteristics of constipated patients according to a positive LMBT

LMBT Negative Positive p-value (n = 42) (n = 17) Age (year) 54.76 ± 14.90 55.82 ± 10.23 0.80 BMI (kg/m2) 21.89 ± 2.05 22.10 ± 2.33 0.74 Presence of rectocele, n (%) Yes (n = 33) 22 (52.4) 11 (64.7) 0.39 No (n = 26) 20 (47.6) 6 (35.3) Total symptom scores 59.39 ± 31.77 60.30 ± 25.67 0.63 Frequency score 31.64 ± 15.54 31.80 ± 13.17 0.68 Bothersome score 27.76 ± 16.97 31.80 ± 13.17 0.58 Anorectal manometry Basal resting pressure, mmHg 72.18 ± 30.58 76.38 ± 23.19 0.61 Maximum resting pressure, mmHg 164.89 ± 61.86 157.32 ± 55.57 0.66 Anal canal length, cm 3.90 ± 1.05 3.48 ± 0.95 0.17 Resting HPZ pressure, mmHg 99.17 ± 37.53 105.82 ± 28.91 0.52 Maximum HPZ pressure, mmHg 143.78 ± 50.37 160.64 ± 44.42 0.23 Basal squeezing pressure, mmHg 105.10 ± 33.82 116.48 ± 39.99 0.27 Maximum squeezing pressure, mmHg 232.35 ± 77.33 239.18 ± 66.14 0.75 HPZ length, cm 1.92 ± 0.50 1.79 ± 0.55 0.38 Minimum sensory volume, mL 62.14 ± 39.73 64.12 ± 37.59 0.86 Urgency volume, mL 96.67 ± 43.04 102.35 ± 49.56 0.66 Maximum tolerable volume, mL 153.33 ± 64.27 154.71 ± 53.52 0.94 Transit time on day 7 (hrs.) Right colon 14.20 ± 15.52 20.96 ± 23.10 0.20 Left colon 19.43 ± 17.74 26.75 ± 16.37 0.15 Rectosigmoid colon 17.69 ± 116.85 20.40 ± 20.49 0.60 Total colon time 51.31 ± 39.23 68.12 ± 43.66 0.16 Types of constipation, n (%) Normal transit (n = 41) 33 (78.6) 8 (47.1) 0.02 Delayed transit (n = 18) 9 (21.4) 9 (52.9) Data are expressed as mean ± SD or number (%). BMI: body mass index (kg/m2); LMBT: lactulose methane breath test; HPZ: high pressure zone.

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017 120 J.-M. Lee et al.

correlation between total CH concentration and the rectosig- Table 3. Multivariate analysis of the predicting factors 4 associated with a positive LMBT moid colon (γ = 0.481, p < 0.01) and total colonic transit time (γ = 0.402, p = 0.02). A maximum HPZ pressure of > 155 mmHg Odds ratio was the only signifi cant independent factor of LMBT+ accord- Variables p value ing to multivariable logistic regression analysis (Table 3). (95% CI) In constipated patients* The presence of rectocele 3.06 (0.70-13.36) 0.14 DISCUSSION Maximum HPZ pressure > 155 mmHg 3.83 (0.93-15.9) 0.06 This study demonstrated that LMBT+, which indicates the Delayed transit 5.52 (1.41-21.74) 0.01 presence of CH4-producing SIBO, was frequent in consti- In rectocele patients pated patients with rectocele. Moreover, breath CH4 is cor- related with an increased rectosigmoid colonic transit time Maximum HPZ pressure > 155 mmHg 8.93 (1.14-71.4) 0.04 and LMBT+ is signifi cantly associated with a high value of Delayed transit 3.45 (0.39-30.30) 0.27 maximum HPZ pressure. *Including the patients with rectocele and functional constipation. HPZ: high pressure zone. The overall rate of LMBT+ in constipated patients with rec- tocele was 33.3% (11/33), in comparison to 23.1% (6/23) in patients with functional constipation. There is no reported Characteristics of rectocele patients according data for rectocele, however, the LMBT+ rate in patients with to a positive LMBT functional constipation in this study was consistent with the fi ndings of previous studies, and the correlation was There were no differences between patients that were LMBT+ 25.8% (16/62) (4). Some researchers believe that jejunal as- and LMBT negative (LMBT-) with regard to age, size of recto- pirate and culture are the gold standard methods to identify cele, colonic transit time parameters or the type of constipa- SIBO. Jejunal aspiration has some limitations, including its tion. With regard to BMI and manometry parameters, basal invasive nature, lack of access to the distal small intestine, squeezing pressure (117.03 ± 32.73 vs 98.15 ± 26.93 mmHg, the potential for contamination during sampling, low repro- p = 0.09), maximum squeezing pressure (247.45 ± 54.88 vs ducibility and the possibility of missing bacteria due to the 207.81 ± 53.24 mmHg, p = 0.06) and maximal HPZ pressure patchy distribution of SIBO (2,17,18). However, the breath (167.31 ± 39.65 vs 131.10 ± 40.30 mmHg, p = 0.02) tended to test is a simple alternative and an acceptable method for be higher or were signifi cantly higher in LMBT+, whereas BMI the diagnosis of SIBO. Currently, the glucose breath test is (20.77 ± 1.59 vs 22.27 ± 1.93, p = 0.03) and HPZ length (1.65 ± widely used to evaluate SIBO (19,20) and this study used 0.43 vs 2.06 ± 0.43 cm, p = 0.02) were lower in LMBT+ patients the LBT. Lactulose is a non-digestible sugar substrate that than in LMBT- patients. However, LMBT+ patients had high passes unabsorbed through the small bowel into the colon. individual total symptom scores in comparison to individu- This might be advantageous in the determination of SIBO, als in the LBMT- group with regard to abdominal discomfort, which is understood to be associated with an impaired in- hard stool, straining during defecation and fl atulence (Fig. testinal motility status, which is expected in constipated 3). There were no signifi cant differences between frequency cases. Glucose is easily absorbed in the proximal small scores, intrusiveness scores, total symptom scores or individ- bowel and rarely reaches the distal small bowel or the co- ual symptoms between the two groups. There was a positive lon (20,21). The North American Breath Testing Consensus published in 2017 stated that both the lactulose and glucose breath tests could be used for the diagnosis of SIBO or to evaluate excessive methane excretion in the breath test in association with clinical constipation (22).

Rectocele is commonly reported in adult female patients with an obstructed defecation syndrome (23). Small recto- celes (< 2 cm) are common even in the healthy population, they can be considered as a normal variant secondary to trauma and advanced age and do not require further treat- ment (17,24,25). Therefore, female patients with rectoceles 2 cm or greater in size were enrolled into the study. It is not clear whether rectoceles are a cause or a result of consti- pation. A recent study suggests that rectoceles are more likely to be a result, rather than the cause of obstructed defecation syndrome (26). This means that surgical pro- cedures intended to correct a rectal wall prolapse are not effective for improving symptoms. In fact, studies of the postsurgical outcome of procedures to correct anatomic abnormalities presumed to be the cause of obstructed def- ecation syndrome are inconsistent (27-30). We assumed that rectoceles could be relevant to the functional problem rath- Fig. 3. Total symptom scores of individual intestinal er than organic disease. Therefore, we analyzed the asso- symptoms according to the positivity of the LMBT in ciation between rectocele in association with constipation constipated patients with rectocele. and LMBT+ in this study.

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017 The association between a positive lactulose methane breath test and rectocele in constipated patients 121

The rectocele patients were divided into two groups influence the results of the breath test. Further studies are (LMBT+ group and LMBT- group) and the characteristics needed due to the limited information for historic controls. of both groups were compared. In terms of BMI and ma- Secondly, there was no information with regard to symp- nometry parameters, squeezing pressure and maximal HPZ tomatic rectocele as well as pelvic outlet dysfunction due pressure were higher in LMBT+ cases, whereas the levels to the limitation in the retrospective design of the study. of BMI and HPZ length were lower in LMBT+ patients in Some patients defined as having functional constipation comparison to LMBT- cases. However, multivariate analysis also show signs of pelvic outlet dysfunction, which could showed that a maximum HPZ pressure of > 155 mmHg was delay colonic transit time. Although the main limitation of the only independent factor of LMBT+. Although the reason the study is the retrospective design, all data were collected is not clear, it is possible that, anatomically speaking, BMI consecutively and prospectively with the same standard could be related to the HPZ length, which mutually act as approach. Demographic data was collected from partici- the confounding variables. HPZ pressure itself may be less pants and they completed a validated intestinal symptom affected by BMI, which was the only independent factor questionnaire and were also asked about their history of identified in the multivariate analysis. Although the pre- drug use or surgical procedures prior to the test in order cise association between LMBT+ and a status of rectocele to avoid recall bias. Strict precautions were taken when with constipation is unclear, a previous study reported that performing the breath and motility tests, irrespective of the

breath CH4 was related to functional constipation, particu- enrollment into the study, in order to obtain an adequate larly in left colonic transit (4). In our study, no significant dif- clinical evaluation. ferences were found in the LHBT profiles among the three groups (i.e., rectocele, functional constipation and control In conclusion, the prevalence of SIBO (which produces

groups). However, the LMBT profiles are the highest at all the CH4 detected in the LMBT) in constipated patients with the time points in the constipated rectocele group and the rectocele is common. Moreover, increased rectosigmoid lowest at all the time points in the control group; the profile colonic transit time or high anorectal pressure might be

of the functional constipation group is in between the two associated with CH4 breath levels detected by the LBT. Fu-

(Fig. 2). A high concentration of breath CH4 might be an ture studies are needed to fully understand the role of intes- important etiological indicator in constipated patients with tinal bacteria in patients with rectocele and to demonstrate rectocele. The proposed pathophysiological mechanism is the potential responses of intestinal bacteria to antibiotic that methanogenic bacteria are predominantly found in the treatment. left colon, which could slow intestinal transit (31-33). In our

study, total breath CH4 concentrations in all the constipated patients (including patients with functional constipation or REFERENCES rectocele) were strongly related with an increased left co- lonic transit time or rectosigmoid colonic transit time, and 1. Schey R, Cromwell J, Rao SS. Medical and surgical management of pelvic delayed transit was an independent factor for a positive floor disorders affecting defecation. Am J Gastroenterol 2012;107(11):1624- 33. DOI: 10.1038/ajg.2012.247 LMBT. On the other hand, in this study, breath CH4 was sig- nificantly associated with increased rectosigmoid colonic 2. Pimentel M, Lin HC, Enayati P, et al. Methane, a gas produced by enteric transit time and independently related to high HPZ pressure bacteria, slows intestinal transit and augments small intestinal contracti-

only in constipated patients with rectocele. Intestinal CH4 le activity. Am J Physiol Gastrointest Liver Physiol 2006;290(6):G1089-95. has been shown to increase the contractile response of the DOI: 10.1152/ajpgi.00574.2004 small bowel in animal models (2). We hypothesize that the 3. Kunkel D, Basseri RJ, Makhani MD, et al. Methane on breath testing is localization and accumulation of CH4 gas in the anatomi- associated with constipation: A systematic review and meta-analysis. Dig cal site of the rectocele herniation in the anal canal might Dis Sci 2011;56(6):1612-8. DOI: 10.1007/s10620-011-1590-5 interact with and delay rectosigmoid colonic transit and increase rectosigmoid contractile activity. These together 4. Lee KM, Paik CN, Chung WC, et al. Breath methane positivity is more com- cause a vicious cycle. mon and higher in patients with objectively proven delayed transit cons- tipation. Eur J Gastroenterol Hepatol 2013;25(6):726-32. DOI: 10.1097/ MEG.0b013e32835eb916 In this study, the LMBT+ group had high individual bowel symptom scores, including a difficulty with defecation-re- 5. Leung L, Riutta T, Kotecha J, et al. Chronic constipation: An evidence-ba- lated symptoms such as abdominal discomfort, hard stool, sed review. J Am Board Fam Med 2011;24(4):436-51. DOI: 10.3122/jab- fm.2011.04.100272 straining during defecation and , in comparison to the LMBT- group. However, there were no significant differ- 6. Longstreth GF, Thompson WG, Chey WD, et al. Functional bowel disorders. ences in individual symptoms between the LMBT positive Gastroenterology 2006;130(5):1480-91. DOI: 10.1053/j.gastro.2005.11.061 and negative groups. Intestinal symptoms such as func- tional constipation, bloating and are potentially 7. Drossman DA. Rome III: The new criteria. Chin J Dig Dis 2006;7(4):181-5. DOI: 10.1111/j.1443-9573.2006.00265.x ameliorated by treatment with antibiotics targeting intesti- nal bacteria, possibly represented by a positive breath test 8. Faccioli N, Comai A, Mainardi P, et al. Defecography: A practical approach. (34,35). These previous studies suggest that antibiotics can Diagn Interv Radiol 2010;16(3):209-16. DOI: 10.4261/1305-3825.dir.2584- improve the symptoms of LMBT+ patients. However, this 09.1 needs to be verified in further studies. 9. Kim EJ, Paik CN, Chung WC, et al. The characteristics of the positivity to the lactulose breath test in patients with abdominal bloating. Eur J There are some limitations in this study. First, the study Gastroenterol Hepatol 2011;23(12):1144-9. DOI: 10.1097/MEG.0b013e- used historical controls. The LBT+ rate in control subjects 32834b0e5c was similar to that of normal subjects (36) and this ap- 10. Park JM, Choi MG, Oh JH, et al. Cross-cultural validation of Irritable Bowel proach has been validated in previous studies (4,9,10). How- Syndrome Quality of Life in Korea. Dig Dis Sci 2006;51(8):1478-84. DOI: ever, age differences between patients and controls may 10.1007/s10620-006-9084-6

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017 122 J.-M. Lee et al.

11. Lee KM, Paik CN, Chung WC, et al. Clinical significance of colonic diverti- 24. Turnbull GK, Bartram CI, Lennard-Jones JE. Radiologic studies of rec- culosis associated with bowel symptoms and colon polyp. J Korean Med tal evacuation in adults with idiopathic constipation. Dis Colon Rectum Sci 2010;25(9):1323-9. DOI: 10.3346/jkms.2010.25.9.1323 1988;31(3):190-7. DOI: 10.1007/BF02552545

12. Rao SS, Azpiroz F, Diamant N, et al. Minimum standards of anorectal ma- 25. Yoshioka K, Matsui Y, Yamada O, et al. Physiologic and anatomic assess- nometry. Neurogastroenterol Motil 2002;14(5):553-9. DOI: 10.1046/j.1365- ment of patients with rectocele. Dis Colon Rectum 1991;34(8):704-8. DOI: 2982.2002.00352.x 10.1007/BF02050355

13. Wald A. Colonic and anorectal motility testing in clinical practice. Am J 26. Hicks CW, Weinstein M, Wakamatsu M, et al. Are rectoceles the cause Gastroenterol 1994;89(12):2109-15. or the result of obstructed defaecation syndrome? A prospective ano- rectal physiology study. Colorectal Dis 2013;15(8):993-9. DOI: 10.1111/ 14. Jung HK, Kim DY, Moon IH, et al. Colonic transit time in diabetic patients - codi.12213 Comparison with healthy subjects and the effect of autonomic neuropathy. Yonsei Med J 2003;44(2):265-72. DOI: 10.3349/ymj.2003.44.2.265 27. Lang RA, Buhmann S, Lautenschlager C, et al. Stapled transanal rectal re- section for symptomatic intussusception: Morphological and functional out- 15. Arhan P, Devroede G, Jehannin B, et al. Segmental colonic transit time. Dis come. Surg Endosc 2010;24(8):1969-75. DOI: 10.1007/s00464-010-0889-1 Colon Rectum 1981;24(8):625-9. DOI: 10.1007/BF02605761 28. Pescatori M, Dodi G, Salafia C, et al. Rectovaginal fistula after double-sta- 16. Metcalf AM, Phillips SF, Zinsmeister AR, et al. Simplified assessment pled transanal rectotomy (STARR) for obstructed defaecation. Int J Colo- of segmental colonic transit. Gastroenterology 1987;92(1):40-7. DOI: rectal Dis 2005;20(1):83-5. DOI: 10.1007/s00384-004-0658-5 10.1016/0016-5085(87)90837-7 29. Zittel TT, Manncke K, Haug S, et al. Functional results after laparoscopic 17. Ford AC, Spiegel BM, Talley NJ, et al. Small intestinal bacterial over- rectopexy for . J Gastrointest Surg 2000;4(6):632-41. DOI: growth in irritable bowel syndrome: Systematic review and meta-analy- 10.1016/S1091-255X(00)80114-6 sis. Clin Gastroenterol Hepatol 2009;7(12):1279-86. DOI: 10.1016/j. cgh.2009.06.031 30. Trompetto M, Clerico G, Realis Luc A, et al. Transanal Delorme procedure for treatment of rectocele associated with rectal intussusception. Tech Co- 18. Sachdeva S, Rawat AK, Reddy RS, et al. Small intestinal bacterial over- loproctol 2006;10(4):389. DOI: 10.1007/s10151-006-0315-9 growth (SIBO) in irritable bowel syndrome: Frequency and predictors. J Gastroenterol Hepatol 2011;26(Suppl 3):135-8. DOI: 10.1111/j.1440- 31. McKay LF, Eastwood MA, Brydon WG. Methane excretion in man - A study 1746.2011.06654.x of breath, flatus, and faeces. Gut 1985;26(1):69-74. DOI: 10.1136/gut.26.1.69

19. Simren M, Stotzer PO. Use and abuse of hydrogen breath tests. Gut 32. Miller TL, Wolin MJ. Enumeration of Methanobrevibacter smithii in human 2006;55(3):297-303. DOI: 10.1136/gut.2005.075127 feces. Arch Microbiol 1982;131(1):14-8. DOI: 10.1007/BF00451492

20. Sellin JH, Hart R. Glucose malabsorption associated with rapid intestinal 33. Nottingham PM, Hungate RE. Isolation of methanogenic bacteria from fe- transit. Am J Gastroenterol 1992;87(5):584-9. ces of man. J Bacteriol 1968;96(6):2178-9.

21. Bond JH Jr, Levitt MD. Use of pulmonary hydrogen (H2) measurements 34. Peralta S, Cottone C, Doveri T, et al. Small intestine bacterial overgrowth to quantitate carbohydrate absorption. Study of partially gastrectomized and irritable bowel syndrome-related symptoms: Experience with Rifaximin. patients. J Clin Invest 1972;51(5):1219-25. DOI: 10.1172/jci106916 World J Gastroenterol 2009;15(21):2628-31. DOI: 10.3748/wjg.15.2628

22. Rezaie A, Buresi M, Lembo A, et al. Hydrogen and methane-based breath 35. Fumi AL, Trexler K. Rifaximin treatment for symptoms of irritable bowel testing in gastrointestinal disorders: The North American Consensus. Am J syndrome. Ann Pharmacother 2008;42(3):408-12. DOI: 10.1345/aph.1K345 Gastroenterol 2017;112(5):775-84. DOI: 10.1038/ajg.2017.46 36. Pimentel M, Chow EJ, Lin HC. Normalization of lactulose breath testing 23. Morandi C, Martellucci J, Talento P, et al. Role of enterocele in the obstruc- correlates with symptom improvement in irritable bowel syndrome. A ted defecation syndrome (ODS): A new radiological point of view. Colorec- double-blind, randomized, placebo-controlled study. Am J Gastroenterol tal Dis 2010;12(8):810-6. DOI: 10.1111/j.1463-1318.2009.02050.x 2003;98(2):412-9. DOI: 10.1111/j.1572-0241.2003.07234.x

REV ESP ENFERM DIG 2018:110(2):115-122 DOI: 10.17235/reed.2017.5017/2017