Obesity among adults in the United States has now reached epidemic proportions. Trends toward obesity are now translating into greater numbers of overweight patients requiring specialized nutrition support and consideration of percutaneous endoscopic gastrostomy (PEG) placement. Due to their safety profile, PEGs are considered the preferred route for long-term nutrition support in those individuals who have a functional gastrointestinal tract.
It has been >2 decades since the original description by Gauderer et al1 of the endoscopically guided procedure now recognized as PEG. Many centers throughout the world have published improvements and favorable results in the debilitated, elderly, adult, children, and infant populations. Complications associated with PEG placement are well described and are usually minor, but major events requiring surgery or resulting in death do occur.2-4 Thirty-day mortality rates range anywhere from 4% to 30%, with pneumonia, urinary tract infections, and heart disease recognized as significant comorbidities.6-7
Traditionally, PEG placement in the overweight and obese patient has been considered a relative contraindication for unclear reasons. Technical challenges include a paucity of anatomical landmarks, impedance of gastric transillumination, and inability to approximate the abdominal and gastric wall. In addition, emphasis has been placed on recognizing and prioritizing the calorically depleted patient as opposed to identifying those who are calorically abundant.
Advances in the treatment of many head and neck cancers, gastrointestinal malignancies, as well as PEG technology are resulting in more favorable results. PEGs are now being placed more frequently in overweight and obese patients with potentially reversible disease states lasting 4 weeks or longer. This is in contrast to traditional PEG placement in the cachetic and terminally ill patients. These findings are not surprising, given the ease and convenience of placement, reversibility, and overall patient satisfaction seen with PEG placement compared with alternative access methods.
As the requirement for specialized nutrition in the overweight patient is expected to grow, so is the need for PEG placement in this population. To date, we are aware of no literature evaluating the overall PEGrelated success rate, morbidity, and mortality in the overweight and obese patient.
PATIENTS AND METHODS
Patients
Consecutive consultation requests for PEG placement in 415 patients were retrospectively evaluated. Patients were managed and followed according to our nutrition support team protocol. Patients were followed over a 1-year period from November 2003 to December 2004. Figure 1 shows the breakdown of the initial consultations. Of these, there were 134 patients that were overweight, with 80 patients in this group meeting criteria for obesity with a body mass index (BMI) s30 kg/m2. This group was composed of 71 males and 63 females. Of the 415 patient consultations for PEG placement, 60 patients were not offered an attempt at PEG placement due to coagulopathy, <30-day projected survival, previous surgery limiting proper access, or obstructing head/neck cancers precluding endoscopic access.
Methods
The Ponsky-Gauderer pull method of PEG placement was attempted on 355 patients either in our endoscopy center or at the bedside over a 1-year period.8 Once patients were appropriately selected, G?* sedation was accomplished with midazolam, meperidine, or sublimaze, as required, and vital signs were monitored. The pull method of PEG placement was the only method employed in this series of patients. Our protocol included a repeat endoscopy to confirm appropriate internal bumper position in all cases. All gastrostomy kits where made of polyurethane and included either 20-Fr PEG kits (BARD, Billerica, MA; or Boston Scientific, Boston, MA) or 24-Fr PEG kits (Wilson-Cook, Winston-Salem, NC). When a jejunal extension was considered or required, a Wilson-Cook 12-Fr jejunal tube was used.
All individuals were evaluated for appropriate antibiotic coverage. Patients who were not currently receiving antibiotics received a single dose of IV antibiotics (usually a first-generation cephalosporin) 20 minutes before the procedure.
Our nutrition support team followed post-PEG nutrition status, procedure-related short- and long-term outcomes, patient complications, and 30-day survival. BMIs were calculated at the time of initial consultation. The type of PEG tube placed, as well as the distances between the internal and external bumpers, was recorded as our standard placement protocol. Feedings were generally started within 24 hours of PEG placement either by continuous pump-infused drip feeding or intermittent infusion administration. Every patient was examined immediately after the procedure, postprocedure day 1, before discharge, and as requested. The mean follow-up period was 24 months.
Analysis and Statistics
An inflammatory response was defined as the following: the presence of a rim of erythema, induration, or the presence of purulent or foul-smelling material. This response may or may not have been accompanied by PEG site pain or local skin infection. Granulation tissue was identified as a soft, easily friable site change with minimal bleeding but without pain, indication of infection, or inflammation.
Data are presented as numbers and percentages or mean
RESULTS
Table I summarizes the underlying clinical conditions and descriptive characteristics of the overweight/obese patients referred for PEG in this study. Patients requiring PEG were classified with either neurologic disease, malignancy, or trauma. During the study period, 355 PEG procedures were attempted. Of those, 134 were found to be overweight, with a BMI ^27 kg/m2, and 80 of those patients were considered obese, with a BMI 2:30 kg/m2. Fourteen out of 355 PEG procedures were considered failures. A failure was defined as an upper endoscopy without subsequent PEG placement. PEGs were aborted secondary to abnormal pathology found during the screening portion of the upper endoscopy and included peptic ulcer disease, severe erosive gastritis, and gastric varices. Of the 14 failures, 1 patient was found to be overweight and 3 were obese. In these 4 patients in whom a PEG was not placed, no trocar was passed. In all 4 instances, no trocar insertion was attempted secondary to a paucity of anatomical landmarks, inability to approximate the gastric and abdominal wall, as well as failure to transilluminate the abdominal wall. Successful gastrostomy placement was achieved in 130 of 134 (97%) of overweight/obese patients (p < .05). In those patients in whom PEG placement was attempted with at least 1 pass of the trocar, 100% success was achieved. Of those overweight and obese patients who received a PEG, no significant differences in complications rates were observed between those patients having a PEG placed at the bedside in an intensive care unit and those who received a PEG placed in the hospital endoscopy suite. The overall procedure-related mortality was 0%. The rate of significant complications in overweight and obese patients remained 0% when compared with those patients with a normal BMI. Figure 2 plots the BMI of the overweight/obese patients with the length of PEG tube required for successful gastrostomy placement. The length of PEG tubing was measured from the internal bumper to the external bumper in centimeters. The increased distances observed between the internal and external bumpers in overweight and obese individuals did not correlate with an increase in post-PEG related complications.
Major and Minor Complications Related to the Procedure
Table II summarizes early and late PEG-related complications in our severely obese patients with BMI >35 kg/m^sup 2^. Out of 355 patients, 29 were found to be severely obese, with a BMI (mean ± SD) of 41.3 ± 4.6 kg/m^sup 2^. One death occurred within 30 days post-PEG placement. An autopsy determined this death to be secondary to an unrelated cardiac event. Three additional unrelated deaths were recognized >30 days post-PEG placement. Three minor complications were encountered, including a local inflammatory response requiring an additional course of oral antibiotics. One patient developed a case of intractable nausea and vomiting requiring conversion of the PEG to PEG/jejunostomy (J). A second patient with persistently high residuals required a relook endoscopy and subsequently improved with medical management.
DISCUSSION
Numerous publications have evaluated the efficacy and safety of PEG placement. Prospective clinical studies have confirmed excellent individual acceptance of this method. PEG placement in the overweight and obese is considered a relative contraindication. Currently, we are unaware of data evaluating the overall success rate, morbidity, or PEG-related mortality in the overweight and obese population.
As the prevalence of obesity has risen in the United States, so has the need for safe and effective specialized nutrition support.9 It is now generally considered that, in patients who have head and neck cancers receiving radiation and chemotherapy, PEG placement has become the standard of care before initiating treatments. In addition to lifelong obesity, many of our patients will present having had their cancers diagnosed and treated at a much earlier stage of their disease. Due to these factors, the prototypical malnourished cancer patient requiring PEG placement is being seen less frequently.
In those patients with malignant diseases resulting in malnutrition, maintenance of adequate enteral nutrition is a major medical goal. However, in those who are calorically abundant or overweight, the importance of adequate nutrition support, including PEG placement, can be overlooked. Printen and colleagues first recognized the importance of nutrition support in this population when they identified an increased incidence of wound infections in those undergoing gastric bypass. Choban et al11 examined the incidence of nosocomial infections in 849 patients undergoing general, urological, vascular, or gynecologic surgical procedures with >4 months of observation. Patients were categorized according to their BMI. A significant increase in the rate of nosocomial infections was recognized in those with elevated BMIs. It was also suggested that obese patients tend to have a more prolonged and protracted hospital stay after surgical procedures. These studies point to the importance of early and aggressive nutrition support, including consideration of PEG placement, even in the overweight patient.
Gastrostomy placement was successful in 130 of 134 (97%) overweight patients, 80 of which were obese, with BMIs ranging 30-63 kg/m2. In those patients in whom PEG placement was attempted with at least 1 pass of the trocar, a 100% success was achieved. The increased distances observed between the internal and external bumpers in overweight and obese individuals did not lead to an increase in post-PEG-related complications. No significant differences in complication rates were observed between those receiving a PEG at the bedside compared with those who received a PEG in the endoscopy suite. The overall procedure-related morbidity and mortality was 0% when compared with those patients with a normal BMI. Fourteen patients failed to receive a PEG, 3 of which were obese, and 1 was overweight. These 4 procedures were aborted due to a paucity of anatomical landmarks and failure to transilluminate the abdominal wall. At no time was a trocar passed in which a PEG could not be placed.
In many instances, there were those patients with a large abdominal girth or skin folds requiring several trocar passes without transillumination. Although this was not optimal, it did not lead to an increase in complications. We have found through our experience that appropriate positioning, below the ribs and left of midline, was a more important prognosticator of success than transillumination of the gastric wall in the obese patient. Emphasis was also placed on the appropriate angle of the trocar and position toward the right shoulder. We also use a single thrust maneuver rather than a slow and gentle entry of the trocar. Traditionally, we are more likely to abort a PEG when we are unable to endoscopically recognize an extrinsic compression with manual pressure. This sign becomes even more important in the obese patient, where gastric transillumination may not be successful. Theoretically, trocar length could be a limiting factor, although this has not been the case in our experience. In general, we are able to fasten a syringe to the trocar for added control and precision when skin folds prevent approximation to the abdominal wall. At times, we have applied more manual pressure to assure approximation of the gastric and abdominal wall.
We recognize that our 0% mortality rate is less than other ranges reported. This may be attributable to our sample size, but we also attempt to limit the placement of PEGs in patients who are less likely to survive 30 days, and we also work closely with our Palliative Care Department. Further review of our PEG-related complications for the 6 months after our formal retrospective evaluation reported in this study does reveal 1 postprocedure death in a very-high-risk patient. This patient was obese and had previously had multiple pulmonary emboli. The PEG procedure required interrupting anticoagulation. The PEG was placed uneventfully, but 10 minutes after the procedure was completed, the patient had a cardiopulmonary arrest and could not be resuscitated. The number of minor PEGrelated complications was not significantly different.
Several studies have identified factors that may predict mortality after PEG placement, including old age, urinary tract infections, and previous aspiration episodes. ^14 In addition, Abuksis et al13 demonstrated increased 30-day mortality in hospitalized patients when compared with outpatients receiving a PEG. Our success rate of 100% is higher than the 89%-95% rate usually reported.14 This is likely due to a strict patient selection criterion that makes every effort to avoid PEG placement in the acutely terminal setting. In addition, we have a nutrition support service that is well trained in recognizing those individuals in whom the risks of PEG would outweigh the potential benefits. We use a meticulous prescreening process, assessing for hemodynamic stability, avoiding those who are febrile or have elevated white blood counts, assuring appropriate antibiotic coverage, and reviewing all radiographic images when appropriate. A nutrition support team follows the post-PEG nutrition status, addresses PEG tube care, and helps initiate appropriate nutrition support.
We believe that PEG placement in the overweight and obese patient can be a technically safe procedure. In those overweight and obese patients who require specialized long-term enteral nutrition support, PEG placement should be considered earlier and more frequently. Our experiences suggest that PEG placement is safe and has a low mortality rate when performed in a center that does a high volume of PEGs, even in obese patients who may be medically debilitated.
