*M1747 Efficacy of an Over-tube for Reducing the Risk of Peristomal Infection After PEG Placement: A Prospective Randomized Comparison Iruru Maetani, Masaki Ikeda, Jun-Ichi Shimura, Takeo Ukita, Yoshihiro Sakai, Masatoshi Yasuda Background: Percutaneous endoscopic gastrostomy (PEG) using the conventional pull method has the drawback of a higher frequency of wound infection because the gastrostomy catheter can become contaminated as it is passed through the oral cavity. Aim: To compare the occurrence rate of peristomal wound infection associated with PEG with and without the use of an over-tube for endoscopic variceal ligation. Methods: Over the last two years, consecutive patients with dysphagia received PEG with the intention of including them in the present study. The patients were randomly assigned to two groups: with over-tube (Group I) or without over-tube (Group II). The peristomal area of the patients was evaluated on a daily basis for 1week after PEG. Erythema and exudate were scored on a scale of 0 to 4, and induration was scored on a scale of 0 to 3; a maximum combined score of 8 or higher, or the presence of pus was criteria for infection as proposed by Jain et al (Ann Intern Med 1987). In each group, cefazolin (2 g/day) for 3 days was given prophylactically. In those patients who had received antibiotics previously, the same antibiotics as previously used were administered after the procedure. All procedures in both groups were performed by two investigators using the pull method (24Fr One-step button). Results: Of the 76 patients enrolled, 38 were assigned to each group. PEG was successful in all patients. Three patients were excluded from each group because of death (1 aspiration pneumonia in Group I; 1 sepsis and 1 suffocation in Group II) within one week of the procedure. Therefore, 73 patients, 37 in Group I and 36 in Group II, were evaluated. There was no signiﬁcant differences between the groups in terms of clinical parameters (age, gender, disease, performance score, mode of previous feeding, and recent antibiotic exposure). The occurrence of peristomal infection within one week of PEG was lower in Group I (with over-tube)(2 vs. 12; p=0.0029). The mean daily combined scores in Group I were also signiﬁcantly lower than those in Group II (p<0.0001), and the median maximum parameter scores in Group I were signiﬁcantly lower than those in Group II (erythema, p=0.0062; induration, p=0.0390; exudate, p<0.0001). There was no procedure-related mortality or clinically important wound infections that required surgical intervention in either group. Conclusion: Use of an over-tube during PEG placement offers a lower risk of peristomal wound infection.
*M1749 Randomized Prospective Comparison of Direct Percutaneous Endoscopic Jejunostomy (DPEJ) vs Percutaneous Endoscopic Gastrostomy with Jejunal Extension (PEG-J) Feeding Tube Placement for Enteral Feeding Mark H. Delegge, Gregory Ginsberg, Steve McClave, James DiSario, Glen Lehman, John Fang Objective: Studies suggest that DPEJ provides more stable jejunal access for longterm enteral feeding and is associated with a lower reintervention rate than PEG-J. This study compares DPEJ and PEG-J feeding tube placement with regard to reinterventions and adverse events (AEs) due to tube migration and failure over 3-6 months. Methods: Patients diagnosed with gastroparesis, chronic pancreatitis, or an increased risk of gastroesophageal reﬂux and aspiration, were offered participation in this ongoing, multicenter study. To date, 14 patients (7 per treatment arm) have been enrolled. Six patients had gastroparesis, 4 had increased risk of reﬂux and aspiration, and 4 had chronic pancreatitis. Thirteen patients (93%) had successful device placement; the unsuccessful DPEJ placement was due to an inability to transluminate. Mean procedure time for the DPEJ and PEG-J groups was 34.1 minutes and 39.6 minutes, respectively. Results: Seven reinterventions occurred, 5 PEG-J and 2 DPEJ. Patients of both treatment arms experienced the following AEs: vomiting (14%), abdominal pain (21%), tube site infection (14%), and pain at tube site (7%). Patients in the PEG-J group had the following tube-related AEs: occlusion (14%), loosening (7%), fracture (7%), kinking (14%), and burning sensation at tube site (7%). Peritube leakage occurred in 1 DPEJ patient. To date, 10 patients have completed the study and 3 have reached the 3 to 6-month follow up. Mean patient satisfaction (using a 10-point scale with1 completely dissatisﬁed and 10 completely satisﬁed) at 1-month follow up for the PEG-J and DPEJ groups was 7.33 and 9.5, respectively. Conclusions: This multicenter study is ongoing, and initial trends suggest that the DPEJ device is associated with shorter procedure time, less AEs, fewer re-interventions, and higher patient satisfaction than the PEG-J device.
*M1748 Predicting Outcome in Patients with Percutaneous Endoscopic Gastrostomy Yakub I. Khan, Ryan Fauble, Michael Piper, Sami Akkary, Catherine Lobocki
*M1750 Prognostic Factors of One-Month Mortality After Percutaneous Endoscopic Gastrostomy and Advantages of the Introducer Technique Mitsuhiro Yata, Masami Terada, Hiroyuki Amano, Sadayo Matsumoto, Yukio Tagashira, Junichi Abe, Noriko Yano, Shigeru Kuwayama
INTRODUCTION: Percutneous endoscopic gastrostomy (PEG) is the most common procedure performed to gain enteral access for feeding purposes in patients with a functional gastrointestinal tract. Thirty-day mortality after PEG placement remains high despite improved guidelines and technical reﬁnement. The purpose of our study was to devise a simple Clinical Parameter Score (CPS) to predict mortality in patients evaluated for PEG placement. CPS was compared to previously studied parameters. DESIGN: We performed a prospective cohort study of 68 consecutive patients receiving PEG in two community hospitals in metropolitan Detroit. Charlson Comorbidity Index, Barthel Index, individual clinical and demographic variables were collected at the time of PEG placement. Patients were followed for 6 months. A cohort survival analysis was performed using the Kaplan-Meier survival method and Cox proportional hazards analysis. RESULTS: 64/68 (94%) patients were available for follow-up. Mortality at 7 and 30 days after PEG were 3.2% and 25.4%, respectively. A CPS composed of 10 parameters (mental status, general appearance, oxygen support, Foley catheter, restraints, pressure sores, tracheostomy, albumin, WBC count and obvious cough) was created as a predictor of early mortality. The median score was 3.0 (range: 0-9.0). Patients with a score $ 4.2 were considered to have a poor clinical status. Median overall survival, as assessed by the Kaplan-Meier method, was signiﬁcantly shorter in patients with a CPS $ 4.2 (p = 0.016, log rank test) than in those with lower scores. Poor mental status was also signiﬁcantly related to a reduced survival (p = 0.048). There was a tendency for decreased survival in patients with a Barthel Index < 70 or a Charslon Comorbidity Index $ 4, but they were not statistically signiﬁcant. Cox proportional hazards analysis showed that only CPS $ 4.2 (RR = 5.52; 95% CI = 1.84 - 16.5) and oxygen support (risk ratio (RR) = 4.49; 95% conﬁdence interval (CI) = 1.09 - 8.88) were predictive factors for death by 30 days after PEG placement (p = 0.001 and 0.034, respectively). Mortality at 30 days was not related to gender, age > 75 years, mental status, or other variables examined. CONCLUSIONS: Short-term mortality in patients undergoing PEG remains high. Improved criteria are needed for better patient selection. Our initial ﬁndings suggest that, CPS is a better predictor of mortality than the other parameters and PEG should be avoided in patients with $ 4.2 CPS score. A larger study is needed to further verify our ﬁndings.
Background: Indications for PEG are dysphagia or anorexia due to neurological impairment, dementia or malignancy. Admittedly PEG is recommended for those who are expected to survive more than one month after PEG placement. Meanwhile one-month mortality is of wide range (9%-42%). Some parameters seem to be associated with this outcome. The aim of this study was to prospectively assess these parameters. Method: Dysphagic or anorexic patients admitted between January 2000 and November 2003 were eligible for this study. All indications (mostly dysphagia) for PEG were discussed and accepted by attending physicians. Enteral nutrition (800-1200 kcal/day)was initiated on the third day after PEG. Prophylactic antibiotics were administered for 3 days. Prognostic factors were prospectively collected and compared between the patients who died within one month after PEG and the rest who survived more than one month. Demographic data; age, gender, Charlson comorbidity index, nutrition-related parameters (albumin, hemoglobin, cholesterol), Barthel index (functional status), serum CRP level(inﬂammation index), positive rate for MRSA and PEG techniques (the introducer technique modiﬁed by Funada or the pull technique) Results: 122 PEGs were placed in 46 men and 76 women. (mean age 78.264.5 years, range 48-102 years of age). Thirteen patients (11%, group A:5 men and 8 women) died within one month after PEG and the rest 109 patients (89%, group B) survived more than one month. Causes of death were pneumonia in 9 cases and others in 4 cases (2 apoplexy, 1 cancer, 1 sepsis). As to demographic data, age and gender:no signiﬁcance, Alb(g/dl); group A vs. group B=2.8 vs. 3.4(p#0.02), Hb(g/ dl); 10.2 vs. 11.4(NS), Chol(mg/dl); 142 vs. 152(NS), CRP(mg/dl); 7.2 vs. 2.2 (p#0.03), Charlson comorbidity index; 4.3 vs. 3.1 (p#0.03), mean Barthel index; less than 10 pt. vs. less than 10 pt.(NS), positive rate for MRSA:0.62 vs. 0.31 (p#0.05) and all 13 cases of group A (1 case placed by the introducer technique and 12 cases by the pull one) vs. the other 109 cases of group B (both 39 cases placed by the introducer technique and 70 cases by the pull one) (p#0.05). Conclusion: As for prognostic factors of one-month mortality after PEG, Charlson comorbidity index ($4), high serum CRP level ($7mg/dl), low serum albumin level (#2.8g/dl) and MRSA positive may be correlated with one-month mortality. Furthermore the introducer technique may reduce a short-term mortality after PEG and it also has the advantage of low cost.
VOLUME 59, NO. 5, 2004