See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/225074214

Which Is More Effective to Prevent Enteral Nutrition-Related Complications,

High- or Medium-Viscosity Thickened Enteral Formula in Patients With
Percutaneous Endoscopic Gastrostomy...

Article  in  Nutrition in Clinical Practice · May 2012

DOI: 10.1177/0884533612442726 · Source: PubMed

CITATIONS READS
5 52

4 authors, including:

Satomi Ichimaru Teruyoshi Amagai

Osaka Saiseikai Nakatsu Hospital Mukogawa Women's University
17 PUBLICATIONS   41 CITATIONS    25 PUBLICATIONS   92 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Satomi Ichimaru on 12 September 2016.

The user has requested enhancement of the downloaded file.

442726 NCPXXX10.1177/0884533612442726Thick
ened Enteral Formula in Patients With PEG / Ichimaru et alNutrition in Clinical Practice
2012

Clinical Research
Nutrition in Clinical Practice
Volume 27 Number 4
Which Is More Effective to Prevent Enteral August 2012 545-552

Nutrition–Related Complications, High- or © 2012 American Society

for Parenteral and Enteral Nutrition
Medium-Viscosity Thickened Enteral Formula in DOI: 10.1177/0884533612442726
http://ncp.sagepub.com
Patients With Percutaneous Endoscopic Gastrostomy?  hosted at
http://online.sagepub.com
A Single-Center Retrospective Chart Review

Satomi Ichimaru, RD1, 2; Teruyoshi Amagai, MD, PhD2;

Maki Wakita, RD, MS2; and Yoshihiko Shiro, MD, PhD3

Abstract
Background: Administration of thickened enteral formula (TEF) through a percutaneous endoscopic gastrostomy (PEG) tube is becoming
a common practice in Japan to prevent enteral nutrition (EN)–related complications. However, what constitutes an adequate viscosity of
TEF remains unclear. The aim of this study was to examine the clinical effects of TEFs with different viscosities administered through PEG.
Methods: The subjects were 50 patients admitted to a single institution who underwent PEG placement. Viscosities of TEFs frequently
administered to the patients were measured, and EN-related complications, nutrition intakes, and clinical outcomes were compared
between high- and medium-viscosity TEFs during the first 2 weeks after TEF feeding initiation. Results: The measured viscosities of
high- and medium-viscosity TEFs were 10,382 ± 931 and 3492 ± 296 mPa·s, respectively. Protein and fluid intakes with TEF were
significantly less in the high-viscosity group. There was no significant difference in EN-related complications, energy intakes, or clinical
outcomes between high- and medium-viscosity TEFs. Conclusion: In this study, high-viscosity TEFs showed no statistical difference in
either EN-related complications or clinical outcomes, in comparison with medium-viscosity TEF. (Nutr Clin Pract. 2012;27:545-552)

Keywords
enteral nutrition; viscosity; gastroesophageal reflux; nausea; vomiting; diarrhea

An administration of thickened enteral formula (TEF), Method 2: Analysis of Patients’ Demographics

instead of liquid formula, is becoming a common practice and Clinical Profiles
in Japan to prevent enteral nutrition (EN)–related complica-
Patients’ demographics and clinical profiles were analyzed
tions, such as nausea, vomiting, and gastroesophageal according to types of EN formulas (Tables 1 and 2).
reflux (GER).1 TEF may prevent these complications
because of its high viscosity. Although higher viscosity is
Method 3: Viscosity Measurement of Formulas
thought to act more on preventing EN-related complica-
tions,2 the viscosity of TEF has been proven in our previous Frequently Used in Patients
study to easily change under clinical situations.3 For these To clarify the viscosities of TEFs used in patients, we mea-
reasons, the adequate viscosity of TEF remains unclear. sured the viscosities by a rotational viscometer TV-22 (Toki
The aim of this study was to examine the clinical effects of Sangyo Co, Ltd, Tokyo, Japan; rotor: 1°34′ × R24, speed: 0.5 rpm,
TEFs with different viscosities administered through the per-
cutaneous endoscopic gastrostomy (PEG) tube.

From the 1Department of Nutrition Management, Kobe City Hospital

Methods
Patients admitted to a single institution, Kobe City Medical Center
West Hospital, who underwent PEG placement between April
2010 and March 2011 were enrolled in this study. All data in
methods 1, 2, and 4 were obtained from patients’ medical records.
Method 1: Analysis of Types of EN Formulas
in Patients With PEG
To analyze which EN formulas were most frequently used in
patients with PEG, we conducted a retrospective chart review.
Organization Medical Center West Hospital, Kobe, Japan; 2Department of
Food Science and Nutrition, School of Human Environmental Sciences,
Mukogawa Women’s University, Nishinomiya, Japan; and 3Department
of Neurology, Kobe City Hospital Organization Medical Center West
Hospital, Kobe, Japan.
Financial disclosure: none declared.
Corresponding Author: Satomi Ichimaru, RD, Department of Nutrition
Management, Kobe City Hospital Organization Medical Center West
Hospital, 2-4 Ichibancho, Nagataku, Kobe, Japan; e-mail: Satomi.
ichimaru@gmail.com.

Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
546 Nutrition in Clinical Practice 27(4)

Table 1. Demographics of the Patients Divided Into 4 Groups by Feeding Formula After Percutaneous Endoscopic Gastrostomy
Placement (Result 1)

TEF-1 TEF-2 Miscellaneous TEF Liquid Formula P Value

Patients, No. (%) 21 (42) 16 (32) 5 (10) 8 (16)  
Age, y (IQR) 83.0 (75.5–90.0) 80.0 (73.5–86.8) 84.0 (77.5–88.5) 77.5 (72.8–84.8) .427
Sex, M/F, No. 16/5 10/6 3/2 2/6 .097
Height, cm (IQR) 156.6 (152.3–160.0) 153.6 (149.0–162.0) 154.0 (146.0–156.6) 145.7 (137.5–165.6) .500
Weight, kg (IQR) 45.0 (39.0–49.4) 42.4 (39.5–51.5) 36.4 (35.6–45.2) 42.0 (31.3–49.1) .358
BMI, kg/m2 (IQR) 17.7 (16.3–20.2) 18.4 (17.4–20.9) 16.6 (15.0–19.0) 18.8 (17.6–19.6) .550

BMI, body mass index; IQR, interquartile range; TEF, thickened enteral formula.

Table 2. Clinical Profiles of the Patients Before Percutaneous Endoscopic Gastrostomy Placement (Result 2)

Miscellaneous Liquid
TEF-1 TEF-2 TEF Formula P Value
Histories of GI surgery, No. (%)  
 All 4 (19.0) 3 (18.8) 2 (40.0) 1 (12.5) .676
  Upper GI tract 1 (4.8) 1 (6.3) 0 (0) 1 (12.5) .811
Comorbidities, No. (%)  
  Hiatal hernia 3 (14.3) 1 (6.3) 1 (20.0) 3 (37.5) .271
 GERD 1 (4.8) 0 (0) 0 (0) 1 (12.5) .500
  Pressure ulcer 3 (14.3) 5 (31.3) 0 (0) 2 (25.0) .390
Treatments  
  Antacids on admission, No. (%) 4 (19.0) 6 (37.5) 0 (0) 2 (25.0) .335
  Motility agents on admission, No. (%) 4 (19.0) 2 (12.5) 0 (0) 2 (25.0) .639
  NPO days before EN initiation, d (IQR) 7.0 (2.0–18.0) 11.0 (2.8–23.0) 3.0 (2.0–8.0) 3.5 (2.0–15.0) .565
  Oral intake after EN initiation, No. (%) 3 (14.3) 2 (12.5) 3 (60.0) 3 (37.5) .080
  PN after EN initiation, No. (%) 21 (100) 16 (100) 5 (100) 8 (100) 1.000
  Duration of PN after EN initiation, d (IQR) 7.0 (6.0–13.0) 5.5 (5.0–8.8) 9.0 (4.5–11.5) 13.5 (5.5–18.3) .373

EN, enteral nutrition; GERD, gastroesophageal reflux disease; GI, gastrointestinal; IQR, interquartile range; NPO, nil per os; PN, parenteral nutrition;
TEF, thickened enteral formula. Antacids include proton pump inhibitors and H2 blockers. Motility agents include metoclopramide, mosapride, itopride,
and rikkunshito.

temperature: 25°C) E-type rotator for 30 seconds. The details
of the methods used were the same as those of our previous
study.3
Method 4: Analysis of Patients’ Clinical Outcomes
Clinical outcomes during the first and second weeks after EN
initiation among the groups fed TEFs, of which viscosities
were measured in method 3, were analyzed. Analyzed param-
eters included the following:
1. (1) Respiratory complications: scores of cough and
sputum where frequency of cough and amount of
sputum were assessed by nurses 3 times a day, and
those expressed as “−/±/+/++” were translated to the
scores of “0/0.5/1.0/2.0”; (2) gastrointestinal (GI)
tract–related complications: vomiting, diarrhea, and
constipation, where the definition of diarrhea is loose
or liquid stool with a frequency of 3 or more times a
day during the study period and that of constipation
is a stool frequency fewer than 3 times per week dur-
ing the study period; (3) prophylactic medication for
these complications: antacids and motility agents;
(4) laboratory data: C-reactive protein (CRP), which
was identified positive when CRP ≥ 2.0 mg/dL, and
percutaneous oxygen saturation (SpO2), which was
identified positive when SpO2 <93%; and (5) body
temperature
2. Nutrition intakes through the enteral and/or paren-
teral route: energy (kcal/d, kcal/kg/d), protein (g/d,
g/kg/d), and fluid (mL/d, mL/kg/d)
3. Clinical outcomes: mortality in hospital, length of
hospital stay, and antibiotics amount for treatment
Prophylactic medications were prescribed by physicians
based on their decisions; antacids included proton pump
inhibitors and H2 blockers, and motility agents included meto-
clopramide, mosapride, itopride, and rikkunshito. CRP and
SpO2 were obtained during the first and second weeks after
EN initiation and were not obtained from all patients because

Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
Thickened Enteral Formula in Patients With PEG / Ichimaru et al 547
the frequency of laboratory tests was not standardized among
the physicians. The number of patients who showed each
event with positive laboratory data at least once a week was
identified (Table 4). Parenteral nutrition (PN) and EN intakes
were assessed during the first and second weeks after EN ini-
tiation, and the daily average intakes of each week were com-
pared between different TEF-fed groups. Antibiotics were
prescribed depending on physicians’ decision according to
results of CRP and body temperature.
This study was approved by the Ethical Committee
of Kobe City Hospital Organization Medical Center West
Hospital.
Statistical Analysis
Statistical analysis was performed by the Mann-Whitney U or
Kruskal-Wallis test using SPSS Statistics version 19 (SPSS,
Inc, an IBM Company, Chicago, IL), and P < .05 was consid-
ered significantly different.
Results
Fifty-three patients underwent PEG placement due to swal-
lowing dysfunction caused by cerebral infarction. The bumper
button-type PEG tubes (Ideal Button; Olympus, Tokyo, Japan)
with 24 Fr-diameter 2.5–4.5 cm in length were placed. Three
patients were excluded from this study because 1 took oral
nutrition of more than 20 kcal/kg/d, 1 died 6 days after PEG
placement, and 1 was discharged on the day after PEG place-
ment. The remaining 50 patients were enrolled to be analyzed
(Figure 1).
Result 1
Each EN formula was assigned by the physician’s decision.
Forty-two patients were fed TEF, and 8 were fed liquid for-
mula. Of the 42 patients fed TEF, 21 were fed one TEF
(defined as TEF-1), 16 were fed another TEF (TEF-2), and
5 were fed miscellaneous TEFs (Figure 1).
Result 2
Demographics of the patients, divided by feeding formulas
into 4 groups, are shown in Table 1. There was no significant
difference in age, sex, height, weight, or body mass index
among the 4 groups.
Clinical profiles before PEG placement were analyzed
(Table 2). The results of the clinical profiles among the 4
groups also showed no statistical significance. To measure vis-
cosity (method 3) and analyze clinical outcomes, we included
the most common TEFs (TEF-1 and TEF-2). The others were
excluded because the number of patients in each group was not
enough for statistical analysis.
Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
Result 3
TEF-1 is a commercially available, ready-to-use (RTU) TEF.
RTU-TEFs, which are packed into pouches that have spouts to
connect to the PEG tube, are administered by squeezing manu-
ally. TEF-2 consists of liquid formula and a thickener (Table 3)
and is administered through a PEG tube using a syringe manu-
ally. To prepare TEF-2, we added the thickener to liquid for-
mula and manually stirred the mixture for 60 seconds,
followed by stabilization for 5 minutes. After stirring again for
30 seconds, the viscosity of TEF-2 was measured.
The measured viscosities of the 2 types of TEF were 10,382
± 931 mPa·s in TEF-1 and 3492 ± 296 mPa·s in TEF-2 (Table
3). The viscosity of TEF-1 was similar to that of tomato paste,
and the viscosity of TEF-2 was similar to that of ketchup.
TEF-1 and TEF-2 were identified as high- and medium-viscosity
TEF in this study, respectively.
Result 4
Clinical profiles and medications for EN-related compli-
cations. There was no significant difference in cough, sputum,
or the number of patients with hyperthermia, diarrhea, constipa-
tion, vomiting, and infection among the patients fed TEF-1 and
TEF-2 during either the first or second week after EN initiation.
The motility agents showed statistical difference in the 2 groups
both in the first and second weeks (P = .001, P = .020).
The other clinical parameters showed no statistical differ-
ence (Table 4). Tube occlusion as a mechanical complication
was not observed either in TEF-1 or TEF-2 groups for 2 weeks
after EN initiation, perhaps because the diameter of the PEG
tube was wider than that of ordinary available enteral tubes.
Nutrition intakes. EN and PN intakes of the patients fed
TEF-1 and TEF-2 were analyzed (Table 5). Because EN is with-
held for at least 24 hours after PEG placement in our institution,
PN was provided for all patients at the time of EN initiation and
followed by a reduction in the amount of parenteral feeding
according to the amount of EN. Protein intake with TEF was
significantly less in patients with TEF-1 in the second week
(44.8 vs 55.9 g/d, P = .033). However, when protein intake was
expressed as per kg/body weight/day, the significant difference
in protein intakes between 2 groups disappeared (Table 5). This
difference, expressed in g/day in protein intake, might have
resulted from the different body weight between TEF-1 and
TEF-2 groups, although the body weights in the 2 groups were
not significantly different. Fluid intake with TEF was also sig-
nificantly smaller in patients with TEF-1 in the first week (439.0
vs 633.0 mL/d, P < .001). The other nutrition intakes showed no
significant difference in either group.
Clinical outcomes. There were no significant differences
in hospital mortality, length of hospital stay, or antibiotics for
treatment between TEF-1 and TEF-2 groups during either the
first or second week after EN initiation (Table 6).
548 Nutrition in Clinical Practice 27(4)

Figure 1. Flow diagram showing patient selection (result 1). BMI, body mass index; EN, enteral nutrition; GERD, gastroesophageal
reflux disease; GI, gastrointestinal; NPO, nil per os; PEG, percutaneous endoscopic gastrostomy; TEF, thickened enteral formula.

Discussion GER, have been reported to be improved with TEF com-

pared with a liquid enteral formula.2,4-10 The results of pre-
The administration of TEF through PEG tubes is spreading vious studies on the efficacy of TEF are summarized in
in Japan.1 EN-related complications, such as diarrhea and Table 7.

Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
Thickened Enteral Formula in Patients With PEG / Ichimaru et al 549

Table 3. Results of the Viscosity Measurement and Nutrition Profiles of TEF-1 and TEF-2 (Result 3)

TEF-1 (High Viscosity) TEF-2 (Medium Viscosity)

Viscosity, mPa·s 10,382 ± 931 3492 ± 296
Energy density, kcal/mL 1.65 0.8
Protein, g/100 kcal 4.0 5.0
Formula profile Ready-to-use TEF Liquid formula
  Trade name PG-soft EJ CZ-Hi
  Selling company Terumo Co Clinico Co, Ltd
Thickener profile (Included in PG-soft EJ) Granule
  Trade name ― Tsururinko for Milk and Liquid Diet
 Manufacturer ― Clinico Co, Ltd
Ratio of formula (mL)/water (mL)/thickener (g) ― 100:25:4
Chemical component Agar, pectin Xanthan gum, carrageenan

TEF, thickened enteral formula. TEF viscosity was measured 5 times, and the results are expressed as mean ± standard deviation.

Table 4. Clinical Complications and Prophylactic Medications During the First and Second Weeks After EN Initiation (Result 4-1)

First Week Second Week

TEF-2 TEF-2
TEF-1 (Medium TEF-1 (Medium
(High Viscosity), Viscosity), (High Viscosity), Viscosity),
  n = 21 n = 16 P Value n = 21 n = 14 P Value
Respiratory tract–related complications
  Cough, score/d (IQR) 0.90 (0.12–2.27) 0.57 (0–2.86) .635 0.72 (0.10–1.73) 1.42 (0.66–2.54) .156
  Sputum, score/d (IQR) 2.43 (1.79–3.21) 2.50 (0.79–2.71) .575 2.00 (1.50–3.00) 2.43 (0.94–3.22) .711
GI tract–related complications, No. (%)
 Vomiting 2 (9.5) 1 (6.3) .721 4 (19.0) 0 (0) .087
 Diarrhea 6 (28.6) 5 (31.3) .862 2 (9.5) 2 (14.3) .669
 Constipation 2 (9.5) 2 (12.5) .776 8 (38.1) 4 (28.6) .567
Prophylactic medication, No. (%)
 Antacids 10 (47.6) 9 (56.3) .608 11 (52.4) 8 (57.1) .785
  Motility agents 11 (52.4) 0 (0) .001 13 (61.9) 3 (21.4) .020
Laboratory data, No. (%)  
  CRP ≥2.0 mg/dL 9 (50.0) 5 (50.0) 1.000 7 (58.3) 6 (60.0) .938
 SpO2 <93% 4 (19.0) 3 (18.8) .982 3 (15.0) 1 (7.1) .491
  Body temperature ≥38.0°C 5 (23.8) 4 (25.0) .934 2 (9.5) 1 (7.1) .808

CRP, C-reactive protein; GI, gastrointestinal; IQR, interquartile range; SpO2, percutaneous oxygen saturation; TEF, thickened enteral formula. Cough,
sputum: nursing records expressed as “–/±/+/++” were translated to points of “0/0.5/1.0/2.0.” Diarrhea was defined as loose or liquid stool with a fre-
quency of 3 or more times a day during the study period. Constipation was defined as having a stool frequency fewer than 3 times per week during the
study period. Antacids: proton pump inhibitors and H2 blockers. Motility agents: metoclopramide, mosapride, itopride, and rikkunshito. CRP and SpO2
were not available for all patients. No. (%): number of patients who showed each event at least once a week.

From the results of these studies, the aim of TEF applica-
tion seems to be to prevent an occurrence of GER. The viscosi-
ties of TEFs in prior studies vary from 900 to 20,000 mPa·s
(Table 7). In the TEFs prepared by adding the thickener to liq-
uid formula, the viscosities were measured under individual-
ized conditions of stirring time and time elapsed since
preparation, as well as the energy densities of the formulas.
However, stirring time and time elapsed since preparation are
considered the most influential factors that affect TEF viscos-
ity, even if at the same concentration of the thickener.3
Therefore, the viscosities of TEFs reported in Table 7 may
differ from each other and may not be able to be compared in
terms of the same clinical outcomes.
The viscosity of an RTU-TEF labeled 20,000 mPa·s was
approximately 10,000 mPa·s as measured in the present study. In
addition, it was reported that the methods for the measurement
of TEF viscosity have not been defined, and the viscosity of
RTU-TEF may change according to different measurement con-
ditions.11 From these observations, measurement conditions
must be defined to examine the efficacy of TEF. Moreover, the
modalities to assess complications were not the same (Table 7),
and the results were not proved by statistical analysis.

Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
 550
Table 5. Nutrition Intakes (Result 4-2)

EN PN EN + PN

TEF-1 TEF-2 TEF-1 TEF-2 TEF-1 TEF-2 P

  (High Viscosity) (Medium Viscosity) P Value (High Viscosity) (Medium Viscosity) P Value (High Viscosity) (Medium Viscosity) Value

First week Energy kcal/d 771.0 (566.5–790.0) 625.0 (548.3–629.0) .174 417.0 (238.5–610.5) 332.0 (155.8–573.0) .550 1091.0 (952.0–1329.5) 988.0 (905.0–1098.0) .118
  kcal/kg/d 16.0 (9.0–20.0) 14.0 (9.0–18.0) .434 9.0 (6.0–17.0) 7.0 (3.0–16.0) .455 26.0 (21.0–34.0) 23.0 (18.0–30.0) .159
  Protein g/d 30.9 (22.7–31.6) 31.3 (27.4–31.4) .889 17.1(10.8–29.0) 18.2 (9.5–28.9) .805 48.0 (35.9–57.4) 50.0 (41.5–60.4) .462
  g/kg/d 0.6 (0.4–0.8) 0.7 (0.5–0.9) .474 0.5 (0.2–0.9) 0.4 (0.1–0.7) .613 1.1 (0.8–1.3) 1.1 (0.8–1.5) .944
  Fluid mL/d 439.0 (280.5–486.5) 633.0 (558.5–799.8) <.001 1129.0 (878.5–1525.0) 932.0 (658.5–1475.3) .276 1596.0 (1315.0–1800.5) 1641.0 (1530.0–1970.0) .462
  mL/kg/d 9.0 (5.0–12.0) 15.0 (11.0–17.0) .006 28.0 (16.0–35.0) 19.0 (15.0–33.0) .477 34.0 (26.0–49.0) 41.0 (31.0–45.0) .615
Second week Energy kcal/d 1120.0 (957.0–1281.0) 1116.5 (841.8–1200.0) .425 0 (0–308.5) 0 (0–126.5) .254 1200.0 (1137.5–1393.0) 1200.0 (900.0–1207.3) .096
  kcal/kg/d 28.0 (20.0–31.0) 25.0 (16.5–27.5) .102 0 (0–9.0) 0 (0–3.0) .508 28.0 (22.5–33.5) 25.0 (19.0–28.5) .098
  Protein g/d 44.8 (38.3–51.2) 55.9 (42.0–60.0) .033 0 (0–11.3) 0 (0–8.1) .499 48.0 (45.0–56.0) 58.2 (45.0–60.0) .188
  g/kg/d 1.1 (0.8–1.2) 1.2 (0.9–1.4) .474 0 (0–0.2) 0 (0–0.2) .634 1.2 (1.1–1.4) 1.3 (1.1–1.4) .614
  Fluid mL/d 1214.0 (606.5–1340.5) 1127.5 (759.0–1308.0) .801 114.0 (0–919.0) 0 (0–491.5) .225 1416.0 (1328.0–1607.0) 1254.5 (1050.0–1586.0) .099
  mL/kg/d 29.0 (15.0–31.0) 24.0 (16.0–31.0) .788 2.0 (0–22.0) 0 (0–15.5) .421 32.0 (29.0–39.0) 30.0 (22.5–40.0) .337

EN, enteral nutrition; PN, parenteral nutrition; TEF, thickened enteral formula. Data are expressed as median (interquartile range). The data on fluid intake in TEF includes additional water.

Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
Thickened Enteral Formula in Patients With PEG / Ichimaru et al 551

Table 6. Comparisons of Clinical Outcomes During the First and Second Weeks After EN Initiation Between 2 Groups: High- and
Medium-Viscosity TEF (Result 4-3)

TEF-1 (High Viscosity) TEF-2 (Medium Viscosity) P Value

Mortality in hospital, No. (%) 1 (4.8) 1 (6.3) .845
Length of hospital stay, d (IQR) 56.0 (24.5–80.5) 46.5 (31.3–63.2) .570
Length of hospital stay after EN initiation, d (IQR) 28.0 (13.0–43.5) 17.5 (13.3–32.5) .208
Discharge within 1 week after EN initiation, No. (%) 0 0 1.000
Discharge within 2 weeks after EN initiation, No. (%) 6 (28.6) 5 (31.3) .862
Antibiotics for treatment, g/d (IQR)  
  First week 0.6 (0–1.8) 0.3 (0–2.9) .849
  Second week 0 (0–1.5) 0 (0–0.6) .361

CRP, C-reactive protein; EN, enteral nutrition; IQR, interquartile range.

Table 7. Effectiveness of the TEFs and Their Preparation and Measurement Conditions

Inada Tabei Kanie Nishiwaki Adachi Shizuku

et al4 et al5 Miyamoto6 et al7 et al8 et al9 et al10 Goda2

 Author n = 15 n = 4a n = 16 n=4 n = 17 n = 15 n = 14 n = 32 n = 15

Measurement of  
TEF viscosity, mPa • s
  By author 800–900 900 900 4100 ND (Pudding-like) 2000
10,000 20,000
  By company 2000 2000  
Effectiveness of TEFs  
 Diarrhea + −b  
 GER + + + +b +b −b − ± +
 Fever + +b  
  Pressure ulcer +  
  Nutrition status +c  
Modality to assess Symptoms Esophageal pH Symptoms Symptoms CT Scintigraphy Esophageal pH Symptoms X-ray X-ray X-ray
complication monitoring monitoring
Statistical analysis − − − +c +b +b +b +b − − −
Type of TEF LT LT LT LT LT LT RTU RTU Barium Barium Barium
Preparation of LT  
  Stirring time ND ND ND 5s ND ND ND ND ND
  Time elapsed since ND ND ND 15 min ND ND ND ND ND
    preparation
Condition of viscosity  
measurement
  Type of viscometer ND B B B ND ND ND ND ND ND ND
  Rotational speed ND ND ND ND ND ND ND ND ND ND ND
  Temperature, °C 20 20 20 ND ND ND 25 25 ND ND ND
Duration of administration, min 56 (20–90) ≤3 30–60 4–8 Bolus Bolus 10 10 5–15 5–15 5–15
Enteral route PEG or NGT NGT PEG or NGT PEG PEG PEG PEG PEG PEG PEG PEG

B, B-type viscometer; CT, computed tomography; GER, gastroesophageal reflux; LT, prepared by adding thickener to liquid formula; ND, not described;
NGT, nasogastric tube; PEG, percutaneous endoscopic gastrostomy; RTU, ready-to-use TEF; TEF, thickened enteral formula. “+” means present and “–”
means absent in statistical analysis.
a
Healthy adults.
b
TEF vs LT.
c
TEF vs before TEF.

In this study, we measured the viscosities of the 2 types of
TEF most used in our institution and compared their complica-
tions, nutrition intakes, and clinical outcomes. Our results sug-
gest that no significant difference existed in complications and
outcomes between the 2 groups. This is different from the
results of previously reported studies in which the methods
were not well enough defined.
From these observations, to our knowledge, the present
study may be the first to examine and analyze the effects of
TEF viscosity on complications and clinical outcomes.

Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
552 Nutrition in Clinical Practice 27(4)
Nausea and vomiting were reported to occur in approxi-
mately 12%–20% of patients receiving EN,12 whereas vomit-
ing occurred in 0%–19% of TEF patients in this study.
Although no universal definition of diarrhea exists, this
complication was reported to occur in 2%–63% of EN-fed
patients.12 In the present study, the incidence of diarrhea was
observed in 10%–31% of TEF patients (Table 4).
In terms of EN-related diarrhea or vomiting with TEF, the
incidence in this study seems similar to or lower than that of
previous studies, although the manner of TEF administration
may be different, and manual preparation described in result 3
could have had a chance of bacterial contamination. The time
required for preparing and feeding RTU-TEF is reported to be
significantly shorter compared with that for liquid formula10;
this seems to be one of the advantages of TEF. However, we
did not compare the preparation times between RTU- TEF and
liquid formula reported in the literatures or thickener-added
TEFs. Further investigation is necessary to clarify the facts.
It is worthy to note that this study has some limitations,
including a single center, a retrospective analysis, a small sample
size of patients, and the short period of the study’s observation,
which are not enough to conclude the effectiveness of TEF.
Long-term prospective studies with larger sample sizes will
be necessary to clarify whether the viscosity of TEF has effects
to prevent EN-related complications.
Conclusion
The viscosities of the 2 types of TEF most commonly admin-
istered in a single institution were measured under defined
conditions and were found to be almost 10,000 (TEF-1) and
3500 (TEF-2) mPa·s. The clinical effects of these TEFs were
analyzed in terms of EN-related complications, nutrition
intakes, and clinical outcomes.
From our observations, it can be concluded that the viscosi-
ties of TEFs must be compared under the same definitive con-
ditions for their effects on clinical outcome.
High-viscosity TEF is thought to be effective as an enteral
formula in Japan. However, our observational results, taken
Downloaded from ncp.sagepub.com at American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.) on December 25, 2015
View publication stats
through a single-center, retrospective chart review, suggest
that high-viscosity TEF appears to have no statistically signifi-
cant advantages in either EN-related complications or clinical
outcomes in comparison with medium-viscosity TEF. Further
studies must examine the clinical effectiveness of TEF.
References
  1. Ichimasa A, Ichimaru S. Review of mechanism, current evidence and prac-
tice of enteral semi-solid formula use to prevent gastroesophageal reflux,
diarrhea, and to improve patient care [in Japanese]. Jomyaku Keicho Eiyo.
2010;25(6):1207-1216.
  2. Goda F. Guidebook for Short Time Gastrostomy Tube Feeding of Semi-
Solid Enteral Nutrient—Towards the Quality of Good Life for Gastrostomy
Patients. Tokyo: Ishiyaku Publishers; 2006.
  3. Wakita M, Masui H, Ichimaru S, et al. Determinant factors of the viscosity
of enteral formula: basic analysis of the thickened enteral formula. Nutr
Clin Pract. 2012; 27:82-90.
  4. Inada H, Kaneda K, Yamagata N. Prevention of aspiration pneumonia in
patients with gastroesophageal reflux using REF-P1 (pectin gel) [in Japa-
nese]. Jomyaku Keicho Eiyo. 1998;20(10):1031-1036.
  5. Tabei I, Kubo H, Yano F, Inada H. The effect of viscosity regulating solu-
tion for enteral nutrition against gastro esophageal reflux [in Japanese].
Nihon Shokaki Geka Gakkai zasshi. 2003;36(2):71-77.
  6. Miyamoto H. Application of nutrition support using half solid diet in the
oldest patients with tube feeding [in Japanese]. Jomyaku Keicho Eiyo.
2009;24:807-809.
  7. Kanie J, Suzuki Y, Iguchi A, Akatsu H, Yamamoto T, Shimokata H. Pre-
vention of gastroesophageal reflux using an application of half-solid nutri-
ents in patients with percutaneous endoscopic gastrostomy feeding. J Am
Geriatr Soc. 2004;52(3):466-467.
  8. Nishiwaki S, Araki H, Shirakami Y, et al. Inhibition of gastroesophageal
reflux by semi-solid nutrients in patients with percutaneous endoscopic
gastrostomy. JPEN J Parenter Enteral Nutr. 2009;33(5):513-519.
  9. Adachi K, Furuta K, Morita T, et al. Half-solidification of nutrient does not
decrease gastro-esophageal reflux events in patients fed via percutaneous
endoscopic gastrostomy. Clin Nutr. 2009;28(6):648-651.
10. Shizuku T, Adachi K, Furuta K, et al. Efficacy of half-solid nutrient for
the elderly patients with percutaneous endoscopic gastrostomy. J Clin Bio-
chem Nutr. 2011;48(3):226-229.
11. Yamaga K, Goda H, Komoto A, et al. Method for the measurement of
physical properties of semisolid enteral nutrition supplements [in Japa-
nese]. Jomyaku Keicho Eiyo. 2011;(26):1247-1253.
12. Boullata J, Nieman L, Guenter P, eds. Complication of enteral nutrition.
In: A.S.P.E.N. Enteral Nutrition Handbook. Silver Spring, MD: American
Society for Parenteral and Enteral Nutrition; 2010:267-307.