Critical Care Evaluation

C

EVALUATION OF A FLEXIBLE FECAL INCONTINENCE MANAGEMENT SYSTEM LINICAL

By Anantha Padmanabhan, MD, Mark Stern, MD, Judith Wishin, RN, BSN, Mari Mangino, RN, Karen Richey, RN, and Mary DeSane, RN, BSN, on behalf of the Flexi-Seal Clinical Trial Investigators Group

Background Management of fecal incontinence is a priority in acute and critical care to reduce risk of perineal dermatitis and transmission of nosocomial infections. Objective To evaluate the safety of the Flexi-Seal Fecal Management System in hospitalized patients with diarrhea and incontinence. Methods A prospective, single-arm clinical study with 42 patients from 7 hospitals in the United States was performed. The fecal management system could be used for up to 29 days. The first 11 patients (all from critical care) underwent endoscopic proctoscopy at baseline; 8 of these had endoscopy again after treatment. The remaining 31 patients (from critical or acute care) did not have endoscopy. Results Rectal mucosa was healthy after use of the device in all patients who had baseline and follow-up endoscopy. Physicians and nurses reported that the system was easy to insert, remove, and dispose of; its use improved management of fecal incontinence; and it was practical, caregiver- and patient-friendly, time-efficient, and efficacious. Skin condition improved or was maintained in more than 92% of patients. Patients’ reports of discomfort, pain, burning, or irritation were uncommon. Adverse events were reported for 11 patients (26%). Death (considered unrelated to study treatment) occurred in 5 patients, 2 patients had generalized skin breakdown, and 1 patient had gastrointestinal bleeding after 4 days of treatment. Conclusions The fecal management system can be used safely in hospitalized patients with diarrhea and fecal incontinence. Additional well-designed, controlled clinical trials may help to measure clinical and economic outcomes associated with the device. (American Journal of Critical Care. 2007;16:384-393)

384

AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 2007, Volume 16, No. 4

www.ajcconline.org

D

iarrhea is a common problem that complicates treatment and may worsen patients’ outcomes. In a prospective study,1 33% of hospitalized patients had fecal incontinence during hospitalization, including more than twice as many patients in the intensive care unit (ICU) compared with acute care units (58% vs 24%). A total of 79% patients had liquid or semiliquid stool, and these patients were 11 times more likely than patients with hard or soft formed stool to have fecal incontinence. Fecal incontinence is an established risk factor for pressure ulcers,2-4 which increase morbidity, mortality, length of hospital stay, and treatment costs.5 It is also a risk factor for transmission of nosocomial infection.6 Therefore, management of fecal incontinence is a priority in ICUs and acute care units.

The first step in management is to identify and, if possible, eliminate the source of the incontinence.7 However, fecal incontinence in hospitalized patients, particularly in intensive care, often is of unknown cause, the result of multiple factors, or the result of another treatment that cannot be discontinued. Therefore, interventions that divert and contain feces often are essential to reduce perineal dermatitis in hospitalized patients. Protection of the skin from exposure to moisture and chemical damage from stool becomes a priority. The Flexi-Seal Fecal Management System (FMS; ConvaTec, Division of E. R. Squibb & Sons, LLC, Princeton, New Jersey) is indicated for the fecal management of patients with little or no bowel control and liquid or semiliquid stool.8 The device consists of a soft silicone cannula approximately 1 m long with an annular silicone low-pressure balloon at the distal end and a pouch flange adapter at the proximal end that connects to the collection bag. The catheter balloon is collapsed to permit nontraumatic insertion into the rectal vault. The balloon is wrapped around the free end of the tubing and has

About the Authors Anantha Padmanabhan is chair of the Department of Surgery at Mt Carmel East Hospital and clinical assistant professor of surgery at Ohio State University in Columbus. Mark Stern is the founder of the Atlanta Academic Research Group in Decatur, Georgia. Judith Wishin is the senior clinical research coordinator for the surgical intensive care unit at the University of Florida, Gainesville. Mari Mangino is a clinical nurse research coordinator at Forum Health in Youngstown, Ohio. Karen Richey is the senior clinical research coordinator at the Arizona Burn Center in Phoenix. Mary DeSane is the operations manager in the Office of Clinical Research at Jersey Shore University Medical Center, Neptune, New Jersey. Corresponding author: Anantha Padmanabhan, MD, FACS, 5965 E Broad St, Suite 250, Columbus, OH 43213 (e-mail: [email protected]). FASCRS,

www.ajcconline.org

a finger pocket for introducing the catheter through the anal sphincter (Figure 1A). The catheter cannula is 23 mm in diameter and collapses to 8 mm in diameter for insertion. The balloon is coated with lubricating jelly and is inserted through the anal sphincter until the balloon is well inside the rectal vault. The balloon is then inflated with 45 mL of water or saline to a diameter of about 53 mm with less than 57 mm Hg pressure supplied by a standard syringe (Figure 1B). The cannula has 2 lumens, each 1 mm in diameter, integrated with the wall. Each lumen is interconnected to its own free length of silicone tubing bifurcating from the main cannula approximately halfway between the proximal and distal ends. The free end of each piece of tubing has a standard Luer valve connector. One lumen is interconnected with the balloon and serves as an inflation-deflation line. The proximal end of the cannula is stretched over a plastic adapter that can be hung from the bed rail and used to attach the collection bag. This study was conducted as part of the development of FMS before the system was marketed in the United States. Our primary objective was to describe the safety of the device in hospitalized patients who were experiencing diarrhea and incontinence. Secondary objectives included evaluations of FMS performance and ease of use.

Management of fecal incontinence is a priority in ICUs and acute care units.

Methods Study Design A prospective, single-arm, descriptive clinical study with 42 patients from 7 hospitals in the United States was done. Each study center had at least one physician as the investigator, and most centers had a registered nurse who was the study

AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 2007, Volume 16, No. 4

385

A

B

Figure 1 A: The catheter for the Flexi-Seal Fecal Management System consists of a soft silicone cannula approximately 1 m long with an annular silicone low-pressure balloon at the distal end and a pouch flange adapter at the proximal end. B: After insertion through the anal sphincter, the balloon is inflated with 45 mL of water or isotonic sodium chloride solution to a diameter of about 53 mm. Reproduced with the permission of E. R. Squibb & Sons, LLC.

coordinator (see “Acknowledgments”). The study protocol was approved by an institutional review board for each study site, and the study was conducted in accordance with the principles of good clinical practice and the ethical standards set forth in the Helsinki Declaration of 1975. Eligibility Criteria At each site, the investigator or study coordinator selected hospitalized patients of legal consenting age for inclusion if the patients had incontinence and liquid or semiliquid stool. Key exclusion criteria included recent rectal surgery, anastomosis of the lower part of the large bowel within the preceding year, and recent cardiac event requiring hospitalization within the preceding 3 months. Each patient (or legally authorized representative) provided written informed consent.

Our primary objective was to describe the safety of the FMS device in hospitalized patients who had diarrhea and incontinence.

Interventions and Assessments Patients could be managed with FMS for up to 29 days, in accordance with the maximum duration of usage indicated in the package insert for the system. The device was removed sooner if an investigator or a study coordinator determined it was no longer clinically indicated. The first 11 patients were recruited exclusively from ICUs, and endoscopic proctoscopy was per-

386

formed to assess the condition of the rectal mucosa before insertion of the FMS. These patients were to have endoscopy again after removal of the device. However, endoscopy was completed at the final visit for only 8 patients; among the other 3 patients, 2 died and 1 had final endoscopy deferred because of the patient’s status. When the endoscopic results and a review of adverse events in the first 11 patients indicated no safety concerns, an additional 31 patients were enrolled from ICUs and acute care sites. Endoscopy was not required in these 31 patients. Caregivers, investigators, and study coordinators recorded all adverse events. At each site, the investigator categorized each adverse event according to severity, presumed cause, and seriousness (ie, if the event was fatal, was life-threatening, or prolonged hospitalization). For study of the performance of the FMS, the following demographic and clinical information was recorded for each patient at baseline: diagnoses, mental response status (alert and responsive, sedated and sluggish, sedated and nonresponsive, comatose, or other), physical response status (responsive to touch or command, responsive to pain only, medication-induced paralysis, or other), consistency of stool (totally liquid, liquid with few solid flecks, or semiliquid), frequency of stool (every 8-12 hours, every 6-8 hours, every 4-6 hours, every 24 hours, or continuous), skin care protocol, and presumed cause of diarrhea. At each site, the investigator or study coordinator also used a nonvalidated scale,

AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 2007, Volume 16, No. 4

www.ajcconline.org

Table 1 Skin condition rating scale

scored from 0 to 4 (Table 1), to rate each patient’s skin condition separately for the perineal and buttocks areas. A digital rectal examination was performed; if fecal impaction was detected, disimpaction was performed and enrollment into the study continued at the discretion of the investigator or study coordinator. The investigator or study coordinator inserted the device and recorded whether it was easy to insert. The nurses (caregivers) directly responsible for care of each patient recorded additional evaluations of FMS performance and tolerability during each day of the study. At each site, the investigator or stud coordinator ensured that all caregivers were instructed on proper use of the FMS and the daily study assessment procedures, and the investigator or study coordinator reviewed the daily caregivers’ assessments. Caregivers recorded whether the device had been expelled from the rectum since the last assessment and, if so, how many times it had been reinserted. Caregivers classified each occurrence of leakage as incidental (nonproblematic, anticipated leakage with a stool management device) or significant (stool leakage around the device). Incidental leakage was categorized further as one of the following: no moisture or stool detected from rectum, detectable clear wetness visually and to touch, small amount of disclosed wetness with odor, source unknown, or other. Significant leakage was categorized further as one of the following: FMS remained in place but stool had soiled skin and bed linens confined to area protected by absorbent pads, FMS remained in place but stool had soiled skin and bed linens extending beyond areas protected by absorbent pads, FMS was expelled with large amounts of stool soilage, or other. Caregivers evaluated skin condition daily by using the skin condition rating scale (Table 1). If a patient was alert and responsive, caregivers asked about the presence of discomfort, pain, irritation, or burning in the rectal area. Caregivers recorded the presence of odor, the presence of obstruction in the device, and collection bag changes. Using a subjective rating scale from 1 (strongly agree) to 5 (strongly disagree), caregivers rated how well the FMS improved control of fecal incontinence. Caregivers used the same scale from 1 to 5 to rate how FMS compared with other fecal management strategies in regards to being time-efficient, caregiver friendly, practical, efficacious, and patient friendly. At the final visit, the investigator or study coordinator removed the device and recorded whether the device was easy to remove and whether it was easy to dispose of.

www.ajcconline.org

Score

Term

Description

0

Normal intact skin

Interpreted as normal unbroken skin with certain variations normal for this person (eg, pigmentation, healed scars, skin depressions)

1

Patchy redness

Slightly red to red skin in patchy areas

2

Extensive redness Slightly red to very red skin covering large without breakdown areas

3

Reddened and blistered without open areas

4

Reddened with open Beginning areas of skin breakdown or areas of skin denudation of skin. Blisters are breaking open at this rating level. This rating includes any bleeding

Very red skin covering moderate to large areas. The blisters are not open areas of skin at this point. This rating should be used even if 1 or 2 blisters can be observed

Statistical Analyses Comparative statistical tests were not performed. Safety summaries included data on all enrolled patients who were treated with the study device. Descriptions of device performance were summarized for the intent-to-treat population, consisting of all enrolled patients who were treated with the study device and had at least one assessment after initial evaluation.

Enrolled

42

Did not retain device >24 h

Intent-to-treat population

38 (90)

Diarrhea resolved Transferred from site Death Discharged from hospital Adverse event Did not retain device Patient choice Flow not well controlled Significant fecal leakage Copious drainage Patient removed device Rectal medication required

Treated for 29 days

4 (10)

12 (29) 6 (14) 5 (12) 4 (10) 2 (5) 3 (7) 1 (2) 1 (2) 1 (2) 1 (2) 1 (2) 1 (2)

0 (0)

Figure 2 Patient disposition. Numbers in parentheses are percentages. Because of rounding, percentages total more than 100.

AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 2007, Volume 16, No. 4

387

Table 2 Demographics and baseline clinical characteristics Baseline valuea (N = 42)

Characteristic Age, mean (SD), y

60.7 (20.4)

Sex Male Female

16 (38) 26 (62)

Location on enrollment Intensive care unit Acute care unit

33 (79) 9 (21)

Mentally alert and responsive

17 (40)

Physically responsive to touch

23 (55)

Cause of diarrhea Medication Infection Disease process Unknown

10 (24) 4 (10) 4 (10) 24 (57)

Stool consistency Totally liquid Liquid with few solid particles Semiliquid

22 (52) 12 (29) 8 (19)

Frequency of stool, h Every 8-12 Every 6-8 Every 4-6 Every 2-4 Continuous; every 0-2

1 0 3 10 28

Comorbid conditionsb Pulmonary problem Cardiovascular problem Metabolic problem Infection Renal problem Neurological problem Multiple trauma Malignancy Pressure ulcer Gastrointestinal problem

23 (55) 17 (40) 16 (38) 13 (31) 9 (21) 9 (21) 6 (14) 5 (12) 4 (10) 2 (5)

(2) (0) (7) (24) (67)

a All values are number of patients (percentage) unless otherwise indicated. Because of rounding, percentages for subcategories may not total 100. b Total exceeds 100% because each patient could have more than a single comorbid condition.

Results Disposition of Patients A total of 42 patients were enrolled in the study. The duration of treatment ranged from 1 to 14 (mean, 5.6) days. A total of 35 patients (83%) retained the device until it was removed by an investigator or a study coordinator. In 4 patients (10%), the device was used for less than 24 hours; the remaining 38 patients (90%) had at least one assessment after the baseline assessment and were included in the intent-to-treat population. In all patients, the device was used for less than 29 days. After the first day, use was discontinued in 12 patients (29%)

388

because of the resolution of diarrhea and in 10 patients (24%) because the patients were transferred to another department or were discharged from the hospital; other reasons for discontinuation are summarized in Figure 2. Baseline Data Baseline demographics and clinical characteristics of study patients are summarized in Table 2. Mean age of patients was 60.7 years, and 62% of patients were female. Most patients (79%) were enrolled from ICUs, 40% were mentally alert and responsive, and 55% were physically responsive to touch. Greater than 90% of patients had stool production continuously or every 2 to 4 hours. Greater than two-thirds of patients had more than one clinical diagnosis. The most commonly identified cause of diarrhea was medication (24%), but the cause was unknown for 57% of patients. Safety The first 11 patients enrolled had endoscopic proctoscopy to examine the rectal mucosa before insertion of the device. Of these 11 patients, 2 died, 1 had endoscopy deferred because of the patient’s status, and 8 had endoscopic proctoscopy at the end of the study after removal of the device. Rectal mucosa was healthy after FMS use for all 8 patients who had baseline and follow-up endoscopy. Figure 3 shows sample results from patient 11 before and after use of the FMS. Adverse events were reported for 11 (26%) of the 42 patients (Table 3). The most frequently reported adverse event was death, which occurred in 5 patients (12%); investigators categorized all deaths as unrelated to study treatment. A total of 2 patients (5%) had generalized skin breakdown with unknown relationship to the study device. Serious adverse events were reported for 7 patients (17%): 5 patients died (see previous paragraph), 1 patient (2%) had fever of unknown relationship to use of the FMS, and 1 patient (2%) had gastrointestinal bleeding. The patient with gastrointestinal bleeding had multisystem disease, had been prescribed anticoagulant and antiplatelet therapies, and had gastrointestinal bleeding at the time of admission to the hospital before use of the FMS. After 4 days of study treatment, the patient had bleeding with ulceration in the lower part of the gastrointestinal tract that was recorded as an adverse event with probable relationship to the study device. Endoscopy after the device was removed showed ulceration, but because the patient was not among the first 11 patients recruited into the study, no baseline endoscopy was available for comparison.

AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 2007, Volume 16, No. 4

www.ajcconline.org

A

B

Figure 3 Representative endoscopic proctoscopy images show healthy rectal mucosa in patient 11 before (A) and after (B) use of the Flexi-Seal Fecal Management System. The device was removed after 5 days of treatment because diarrhea had resolved.

Device Performance Investigators and study coordinators reported that the FMS was easy to insert for 37 (97%) of 38 patients. At the final visit, ease of removal and disposal were not rated for 2 patients who had died, and ease of removal was not rated for 1 additional patient who could not retain the device. For the remaining patients, investigators and study coordinators rated FMS easy to remove for 34 (97%) of 35 patients and easy to dispose of for 36 (100%) of 36 patients. Caregivers completed a total of 200 daily reports of device performance (Table 4). On 83% to 90% of these reports, caregivers said they agreed or strongly agreed with each of the following statements: the device improved fecal incontinence control, and it was practical, caregiver- and patient-friendly, timeefficient, and efficacious. Caregivers also reported that odor and blockage were uncommon (15% and 16%, respectively) and reported changing the collection bag at 26% of daily assessments. Caregivers recorded the extent of leakage on 198 daily assessments; most of these reports (83%) indicated minimal or no leakage (Table 5). Of the 39 patients who had evaluations of skin condition at baseline and final visit, 26% had normal skin on the buttocks and 23% had normal skin on the perineum at baseline (Table 6). By the final visit, skin condition was improved or maintained on the buttocks for 92% of patients and on the perineum for 92% of patients.

www.ajcconline.org

On 130 (65%) of the 200 daily assessments, caregivers reported that patients were nonresponsive. Of the 70 daily caregiver reports that included patients’ assessments of physical symptoms, 52 (26%) indicated that patients did not have discomfort, pain, burning, or irritation. Pain was reported at 8 (4%) assessments and irritation at 1 (