- Top of page
Aim: The aim of this study was to clinically evaluate percutaneous endoscopic gastrostomy (PEG) tube feeding of elderly Japanese patients with dementia.
Method: The records of the 155 patients with dementia who underwent PEG in Juntendo Tokyo Koto Geriatric Medical Center were reviewed for pertinent clinical data, including diagnosis of dementia, place of stay before and after hospitalization, as well as survival rate, albumin levels, and incidence of aspiration pneumonia (AP) before and 6 months after PEG feeding. The latter three data of these patients were compared with those of 106 patients with dementia fed through a nasogastric (NG) tube.
Results: Alzheimer's disease and vascular dementia were predominant. Fifty-three percent of the patients were admitted from their home; the number of discharges to homes decreased to 21.2%. The mean (SD) of the albumin levels was 2.9 (0.4) g/dl before feeding and 2.9 (0.6) g/dl after 6 months. Among the patients with AP before PEG tube feeding, 51.6% had an AP recurrence. Conversely, AP occurred in 9.4% of the patients without AP before feeding. The patient survival rate was higher by 27 months when using PEG tube than when using an NG tube.
Conclusion: PEG tube feeding in patients with dementia leads to preservation of status for a few years. Compared with NG tube feeding, PEG tube feeding did not induce AP due to impairment of intact swallowing function, and was associated with higher survival rate of approximately 2 years. However, PEG tube feeding does not seem to promote home medical care.
IN 1980, GAUDERER et al.1 first reported percutaneous endoscopic gastrostomy (PEG), and since then, PEG tube has been used widely in artificial feeding. Although PEG was first introduced as an operation to provide enteral nutrition to children and young adults,1 it was recently recognized as the main method of artificial feeding for elderly patients with chronic or degenerative diseases of the heart, lungs, kidneys, or nervous system, and especially for those with severe dementia.
The process of population aging among the Japanese has been accelerated since the 1990s, and dementia has become a common disease in elderly Japanese people.2 Hence, a method of treatment and care for elderly patients with severe dementia came to be highly recommended. Therefore, PEG tube feeding has been evaluated as a safety method of providing nutrition and a prophylaxis of undernutrition and aspiration pneumonia (AP).3,4 In Japan, more elderly patients are undergoing PEG compared with those in Europe and the USA, and an increase in the use of the operation is expected.3
To evaluate PEG tube feeding of elderly Japanese patients with dementia, we studied the clinical features of patients with dementia who underwent PEG at Juntendo Tokyo Koto Geriatric Medical Center (GMC).
- Top of page
The survey was conducted during the period from June 2002 (the opening of the GMC) to December 2010. A total of 155 patients with dementia who underwent PEG (PEG group: 85 men and 66 women; mean age, 79.1 years) were examined. Clinical features, including diagnosis of dementia by the ICD-10,5 place of stay before and after hospitalization, and survival rate, were investigated, as well as the albumin level and incidence of AP before and 6 months after PEG feeding. The results of the investigation on the survival rate, albumin level, and incidence of AP were compared with those of 106 patients with dementia who were fed through a nasogastric tube (NG tube group: 51 men and 55 women; mean age, 79.4 years) in the same period.
We compared continuous variables using the paired Student's t-test and compared categorical variables using Fisher's exact test. The survival rate was analyzed using the Cutler–Ederer method.6P < 0.05 was regarded as significant.
Informed consent for the use of data was obtained verbally from the patients or from their caregivers according to the procedures approved by the institutional review board of GMC.
- Top of page
The diagnoses for the PEG group are shown in Table 1. Cases with dementia in Alzheimer's disease or vascular dementia, a common dementia in Japan, were predominant. Furthermore, the other types of dementia associated with other conditions, such as progressive supranuclear palsy, dentatorubral–pallidoluysian atrophy, and cortico–basal degeneration, are included in F02.8 (the cases of dementia in other specified diseases classified elsewhere).
Table 1. Diagnoses for the percutaneous endoscopic gastrostomy group by ICD-10 (n = 151)
| ||No. of patients||%|
The patient's places of stay before and after hospitalization are shown in Table 2. More than half of the patients were admitted from their home (53.0%); however, the number of patients who were discharged and sent home from the GMC decreased to 21.2%. On the other hand, 15.9% of the patients were admitted from other hospitals, and the number of patients sent to other hospitals increased to 33.8%.
Table 2. Place of stay before and after hospitalization (n = 151)
| ||Before hospitalization||After hospitalization|| P-value|
|No. of patients||%||No. of patients||%||(Fisher's exact test)|
The albumin levels were measured in 38 patients in the PEG group and 31 patients in the NG tube group (Fig. 1). The mean (SD) value in the PEG group was 2.9 (0.4) g/dl before feeding and 2.9 (0.6) g/dl after 6 months, and there were no significant differences (P = 0.70). In the NG tube group, the mean (SD) albumin value was 2.6 (0.5) g/dl before feeding and 2.9 (0.5) g/dl after 6 months, and similar to the PEG group, there were no significant differences (P = 0.08). Compared with the value 6 months after feeding, no significant differences were obtained between the PEG and NG tube groups (P = 0.84).
Figure 1. Albumin levels before feeding and after 6 months. There were no significant differences between the mean values in the percutaneous endoscopic gastrostomy (PEG) group (darker bars) before and after 6 months' feeding, between mean value of nasogastric (NG) tube group (lighter bars) before and after 6 months' feeding, and between the mean values in each group after 6 months' feeding.
Download figure to PowerPoint
The incidence of AP before and 6 months after feeding was confirmed in 96 patients in the PEG group and 69 cases in the NG tube group (Table 3). In the PEG group, AP occurred in 64 patients before feeding and recurred in 33 of these patients (51.6%) after feeding. On the other hand, of 32 patients without AP before feeding, AP occurred in three (9.4%) only after feeding. Of 17 patients in the NG tube group without AP, AP occurred in nine (52.9%) after feeding as opposed to the PEG group.
Table 3. Incidence of AP before and after 6 months of tube feeding
| || || ||6 months after|| P-value|
|AP (+)||AP (−)||(Fisher's exact test)|
|No. of patients||No. of patients|
|PEG group (96 patients)||Before feeding||AP (+)||33||31||0.90|
|NG tube group (69 patients)||Before feeding||AP (+)||45||7||<0.01|
Using the Cutler–Ederer analysis, we constructed a table of life events and represented this graphically (Fig. 2). The survival rate of the PEG group is significantly higher by 27 months than that of the NG tube group (mean [SD]: PEG group, 65.6 [5.6%]; NG tube group, 44.4 [9.8%]; P = 0.019).
Figure 2. The Cutler–Ederer curve of the percutaneous endoscopic gastrostomy (PEG) and nasogastric (NG) tube groups. The survival rate of the PEG group (solid line) is significantly higher by 27 months than that of the NG tube group (broken line).
Download figure to PowerPoint
- Top of page
In our study, PEG tube feeding was effective in the preservation of albumin; however, a major improvement in albumin levels was not confirmed. Furthermore, the preventive effect of AP was not confirmed in the patients with dysphagia before feeding. Compared to NG tube feeding, PEG tube feeding did not induce AP due to impairment of intact swallowing function, and it was associated with a higher survival rate of approximately 2 years.
According to the guidelines compiled under the supervision of the Japan Gastroenterological Endoscopy Society, patients who cannot feed voluntarily because of cerebrovascular disease or dementia undergo PEG.3 However, in the previous study, the ineffectiveness of the PEG tube in the prevention of AP and in the improvement of nutritional status was reported outside Japan.4,7–9 The guidelines of the European Society for Clinical Nutrition and Metabolism contradicted these findings on the effect of PEG for improving functional status, avoiding hunger, improving comfort, preventing nutritional decline and its consequences, preventing aspiration, and reducing the incidence of pressure sores and infections in elderly patients with dementia.10 Based on the results of the previous study, it is suggested that the effect of PEG tube feeding on the patients with dementia leads to a preservation of status for a few years, and shows no marked improvement even in the albumin levels. Given that hypoalbuminemia is a risk factor for pressure sores and pressure sores with hypoalbuminemia are resistant to treatment,11,12 it appears difficult to improve pressure sores significantly with PEG tube feeding. Preservation of status is important in geriatric treatment; therefore, it could not be determined whether PEG is ineffective for patients with dementia. Accordingly, informed consent regarding the limits of treatment for the patients and their families is necessary before the operation. Kosaka et al.13 reported that 33% of the families of elderly patients, in whom oral ingestion was impossible, refused tube feeding; even after providing sufficient informed consent, 73% of families refused to cooperate.
The Japanese health-care system has had reforms since the 2000s, such as reduction in the number of long-term-care beds and shortening of the average hospital stay.14 As a result of the reform, home medical care was promoted and PEG tube feeding was introduced as a safety method.3 However, the present study showed that most of the patients were discharged to an institution or another hospital after hospitalization and PEG tube feeding did not appear to promote home medical care. Because of the dependence on hospital care, increasing numbers of members of the nuclear family, aging of the caregiver, and capacity of the institution, we indicated that elderly patients with dementia have difficulty being discharged to their home.15 Care of patients who need treatment for physical complications, including the placement of a PEG tube, is especially difficult for elderly caregivers and for the staff of institutions for day service or short stay without the necessary equipment. It was observed that these problems decreased discharges to the home and increased changing of hospitals, which is in contradiction to the reform made in the health-care system. Nursing education for caregivers and the staff and maintenance of the care environment in the area are necessary before introducing high-quality medical care.
Recently, a major Japanese newspaper publisher reported articles regarding challenges associated with PEG in elderly patients, and a few medical personnel questioned the excessive use of PEG.16,17 Because of the further increase in the number and the aging of patients with dementia, it is necessary to collect information about the effects of PEG in Japan.