Patient Services/ NGT Placement Confirmation/ BESt 024

Best Evidence Statement (BESt)

Date published/posted: April 27, 2009

Confirmation of Nasogastric Tube (NGT) Placement

Clinical Questions
P (population/problem) In pediatric and adolescent patients who require an NGT,
I (intervention) are multiple non-radiological verification methods (auscultation and aspiration
methods)
C (comparison) compared to radiological verification methods,
O (outcome) as accurate in confirming NGT placement?

P (population/problem) In pediatric and adolescent patients who require an NGT,

I (intervention) are gastric aspirates, obtained under clinical conditions (i.e. patients who are fed
or fasting, on or off acid-suppression medication), with a pH<6,
C (comparison) compared to a pH<5,
O (outcome) more accurate in confirming NGT placement?

P (population/problem) In pediatric and adolescent patients who require NGT placement,

I (intervention) are tube length predictions using age-related height-based (ARHB) methods,
C (comparison) compared to nose-ear-xiphoid (NEX) morphological measurements,
O (outcome) more accurate in predicting NGT length?

Target Population Neonatal, pediatric and adolescent patients who require NGT for feeding or gastric
decompression

Recommendation(s) (See Table of Recommendation Strength following references)

1. It is strongly recommended that NG tube length be predicted as follows:
a. In patients ≤ 8 years 4 months of age, use age-related height-based (ARHB) methods. (See Table: Age-related
height-based equations for NGT length predictions)
b. In neonates, patients >8 years 4 months of age, patients with short stature, or if unable to obtain an accurate
height, use morphological measurements.
 Age-related height based (ARHB) methods are more accurate than nose-ear-xiphoid (NEX) and other
morphological measurements in predicting tube length for NG placement in children ≤ 8 years 4 months
of age (Beckstrand 2007 [4a], Strobel 1979 [4a], Ellett 1992 [4b]).
Note: When the enteral tube is inserted to the predicted distance using age-specific height regression
equations, the relative percent placement error is less than morphological measurement predictions for
children <8 years 4 months of age (Beckstrand 2007 [4a]).
 Morphological measurements frequently under-predict tube length. The most accurate morphological
measures include nose-ear-mid-xiphoid-umbilicus (NEMU) predictions for patients >8 years 4 months of
age, those with short stature or if unable to obtain an accurate height (Beckstrand 2007 [4a], Ellett 1992 [4b]).
 For neonates the evidence is limited for best morphological measurement (Weibley 1987 [3a], Beckstrand 2007
[4a], Gallaher 1993 [4a], Tedeschi 2004 [4b]).
Note: Measurements using NEMU method for tube length predictions versus NEX method alone in
neonates is slightly more reliable for length prediction (Weibley 1987 [3a], Beckstrand 2007 [4a], Gallaher 1993
[4a]).
 Tube length is best indicated with an indelible marker and recorded in patient documentation by amount
of tube remaining from the nares to the end of the tube (Weibley 1987 [3a]).

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 1 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Table: Age-related height-based equations for NGT length predictions

Predicted distance to
Route Age group
stomach
Nasal 2 weeks to ≤ 28 months 17.6 cm + 0.197 (height cm)
Nasal 28 months to < age ≤8years 4 months 21.1 cm + 0.197 (height cm)
(Beckstrand, 2007 [4a], Strobel, 1979 [4a])

2. It is strongly recommended that multiple non-radiological verification methods be used to confirm placement of
an NGT in neonatal, pediatric and adolescent patients. These methods include:
a. Gastric auscultation: Auscultation as a verification method is 60%-80% reliable (Ellett 1999 [4a], Metheny 1990
[4a], Neuman 1995 [4b], Local consensus [5]). (See Appendix 1 Likelihood ratios for auscultation)
b. Aspirate pH testing: Testing of aspirate pH is the only aspirate method currently available for bedside
assessment. Use an aspirate pH<6 to confirm NGT placement for neonatal, pediatric and adolescent patients,
when obtained under clinical conditions that include fed or fasting patients and those on and off acid-
suppression medications (Ellett 2005 [3a], Metheny 2002 [3a], Metheny 1994A [3a], Phang 2004 [3b], Westhus 2004 [3b],
Ellett 1999 [4a], Metheny 1999A [4a], Metheny 1999B [4a], Nyqvist 2005 [ 4b], Neumann 1995 [4b], Local Consensus [5]). (See
Appendix 2 Likelihood ratios for aspirate pH <6 vs. pH <5 to confirm gastric placement)
 Gastric aspirate pH means are statistically significantly lower compared with means from
intestinal and respiratory pH aspirates (Ellett 2005 [3a], Metheny 2002 [3a], Metheny 1994A [3a], Phang 2004
[3b], Westhus 2004 [3b], Metheny 1999A [4a], Metheny 1999B [4a]). (See Appendix 3 Mean Concentration for
gastric aspirate pH, intestinal aspirate pH and respiratory samples with standard deviation)
 Mean values for aspirate pH are not significantly different when patients are fed or fasting (Metheny 2002
[3a], Metheny 1999B [4a]).
 Mean values for aspirate pH are not significantly different when patients are on or off acid-suppression
medications (Metheny 2002 [3a], Metheny 1999A [4a]).
 Mean values for aspirate pH are not significantly different by age of patient (Ellett 2005 [3a], Metheny 2002
[3a], Metheny 1999A [4a]).
Note: pH strips are as accurate as a pH meter for testing aspirate pH (Ellett 2005 [3a], Metheny 1999A {4a],).
c. Visual inspection of aspirate: Visual inspection is less accurate than pH to confirm placement. Use
in addition to testing aspirate pH. Aspirate colors are specific to the intended placement location
(Metheny 2002 [3a], Metheny 1994A [3a], Phang 2004 [3b], Westhus 2004 [3b], Metheny 1999B [4a], Metheny 1994B [4a]). (See
Appendix 4 Visual characteristics for gastric aspirates)
d. Aspirate testing of enzyme levels for bilirubin, pepsin and trypsin: highly accurate but limited to
laboratory assessment (Ellett 2005 [3a], Metheny 2002 [3a], Westhus 2004 [3b], Metheny 1999A [4a], Local Consensus [5]).
(See Appendix 5 Likelihood ratios for aspirate bilirubin and CO2 to confirm gastric placement and
Appendix 6 Mean pepsin and trypsin concentrations for gastric and intestinal aspirates)
e. CO2 monitoring: CO2 monitoring is another reliable method but requires a capnograph monitor and is used to
determine incorrect tube placement in the respiratory tract (Ellett 2005 [3a], Metheny 2002 [3a], Metnehy 1999A [4a]).
(See Appendix 5 Likelihood ratios for aspirate bilirubin and CO2 to confirm gastric placement).
Note: When aspirate and non-aspirate verification methods are used in combination to confirm NGT
placement, the post-test probability for accuracy increases to 97-99%, approaching the radiological gold
standard of 99% (Ellett 2005 [3a], Metheny 2002 [3a], Metheny 1994A [3a], Metheny 1993 [3a], Phang 2004 [3b], Westhus 2004
[3b], Ellett 1999 [4a], Metheny 1999A [4a], Metheny 1999B [4a], Neumann 1995 [4a], Metheny 1994B [4a]). (See Appendix 7
Likelihood ratios for aspirate color and combined aspirate color and Appendix 8 Algorithm: Confirmation of
Nasogastric or Orogastric (NG/OG) Tube Placement)

3. It is strongly recommended that radiological verification is used when non-radiological methods are conflicting or
patients are considered high risk which include:
a. Patients in pediatric and cardiac intensive care units.
b. Patients exhibiting an altered level of consciousness.
c. Patients with swallowing problems.
(Ellett 2005 [3a], Metheny 1994 [3a], Phang 2004 [3b], Ellett 1999 [4a], Neumann 1999 [4b], Local Consensus [5])

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 2 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Discussion/summary of evidence
There is good evidence that the combination of multiple verification methods to confirm NGT placement will reduce
the required number of X-rays in patients with NGTs (Ellett 2005 [3a], Metheny 2002 [3a], Metheny 1994A [3a], Phang 2004 [3b],
Westhus 2004 [3b], Metheny 1999A [4a], Metheny 1994B [4a], Metheny 1990 [4a], Metheny 1999B [4b], Neumann 1995 [4b]). There is
also good evidence that improving the accuracy of predicting NGT length prior to insertion will enhance the precision
of successful NGT placement (Weibley 1987 [3a], Beckstrand 2007 [4a], Gallaher 1993 [4a], Strobel 1979 [4a], Tedeschi 2004 [4b],
Ellett 1992 [4b]) and potentially reduce healthcare costs (Local Consensus [5]). However, there is limited evidence on the
most accurate tube length predictions for neonates, patients >8 years 4 months of age, patients with short stature or if
unable to obtain an accurate height.

Side Effects and Risks

Although the procedure of aspiration of gastric or intestinal contents is generally well-tolerated, increased tube clogging
and tissue irritation from aspiration can occur (Metheny 1994A [3a], Ellett 1999 [4a]). When considering the pH<6 as a cutoff
for NGT placement confirmation, including clinical conditions of fed or fasting and acid-suppression medications, there is
a greater potential for false negatives which may increase the number of required radiological studies for this population
(Ellett 2005 [3a], Metheny 2002 [3a]). A potential side effect of using age-based height regression methods to predict NGT
length is using inaccurate height/length measurements (Beckstrand 2007 [4a], Ellett 1992 [4b]). Incorrect height/length
measurements can result in NGT length predictions that are too long or too short for individual patients.

References/citations (evidence grade in [ ]; see Table of Evidence Levels and Table of Recommendation Strength following references)
1. Beckstrand, J., Ellett, M. L. C., & McDaniel, A. (2007). Predicting internal distance to the stomach for positioning nasogastric and
orogastric feeding tubes in children. JAN: Journal of Advanced Nursing, February, 274-289. [4a]
2. Ellett, M. L. C., & Beckstrand, J. (1999). Examination of gavage tube placement in children. Journal of the Society of Pediatric Nurses,
4(2), 51-60. [4a]
3. Ellett, M. L. C., Beckstrand, J., Welch, J., Dye, J., & Games, C. (1992). Predicting the distance for gavage tube placement in children.
Pediatric Nursing, 18(2), 119-23-127. [4b]
4. Ellett, M. L. C., Croffie, J. M. B., Cohen, M. D., & Perkins, S. M. (2005). Gastric tube placement in young children. Clinical Nursing
Research, 14(3), 238-52. [3a]
5. Gallaher, K. J., Cashwell, S., Hall, V., Lowe, W., & Ciszek, T. (1993). Orogastric tube insertion length in very low birth weight infants.
Journal of Perinatology, 13(2), 128-31. [4a]
6. Local Consensus during BESt development time frame. [5]
7. Metheny, N., McSweeney, M., Wehrle, M., & Wiersema, L. (1990). Effectiveness of the auscultatory method in predicting feeding tube
location. Nursing Research, 39(5), 262-7. [4a]
8. Metheny, N., Reed, L., Berglund, B., & Wehrle, M. A. (1994B). Visual characteristics of aspirates from feeding tubes as a method for
predicting tube location. Nursing Research, 43(5), 282-7. [4a]
9. Metheny, N. A., Clouse, R.E. Clark, J.M., Reed, L., Wehrle, M. A., & Wiersema, L. (1994A). Techniques & procedures. pH testing of
feeding-tube aspirates to determine placement. Nutrition in Clinical Practice, 9(5), 185-90. [3a]
10. Metheny, N. A., Eikov, R., Rountree, V., & Lengettie, E. (1999B). Indicators of feeding-tube placement in neonates. Nutrition in
Clinical Practice, 14(6), 307-14. [4a]
11. Metheny, N. A., & Stewart, B. J. (2002). Testing feeding tube placement during continuous tube feedings. Applied Nursing Research,
15(4), 254-8. [3a]
12. Metheny, N. A., Stewart, B. J., Smith, L., Yan, H., Diebold, M., & Clouse, R. E. (1999A). pH and concentration of bilirubin in feeding
tube aspirates as predictors of tube placement. Nursing Research, 48(4), 189-97. [4a]
13. Metheny, N., Reed, L., Wiersema, L., McSweeney, M., Wehrle, M., & Clark, J. (1993). Effectiveness of pH measurements in
predicting feeding tube placement: An update. Nursing Research, 42(6), 324-331. [3a]
14. Neumann, M. J., Meyer, C. T., Dutton, J. L., & Smith, R. (1995). Hold that X-ray: Aspirate pH and auscultation prove enteral tube
placement. Journal of Clinical Gastroenterology, 20(4), 293-295. [4b]
15. Nyqvist, K. H., Sorell, A., & Ewald, U. (2005). Litmus tests for verification of feeding tube location in infants: Evaluation of their clinical
use. Journal of Clinical Nursing, 14, 486-495. [4b]

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 3 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

16. Phang, J. S., Marsh, W. A., Barlows, T. G. I., & Schwartz, H. I. (2004). Determining feeding tube location by gastric and intestinal pH
values. Nutrition in Clinical Practice, 19(6), 640-4. [3b]
17. Strobel, C., Byrne, W., Ament, M., & Euler, A. (1979). Correlation of esophageal lengths in children with height: Application to the
tuttle test without prior esophageal manometry. Journal of Pediatrics, 94(1), 81-84. [4a]
18. Tedeschi, L., Altimier, L., & Warner, B. (2004). Improving the accuracy of indwelling gastric feeding tube placement in the neonatal
population. Neonatal Intensive Care, 16(1), 16-18. [4b]
19. Weibley, T. T., Adamson, M., Clinkscales, N., Curran, J., & Bramson, R. (1987). Gavage tube insertion in the premature infant. MCN:
The American Journal of Maternal/Child Nursing, 12, 24-27. [3a]
20. Westhus, N. (2004). Methods to test feeding tube placement in children. MCN: The American Journal of Maternal/Child Nursing,
September/October, 283-291. [3b]

Note: Full tables of evidence grading system available in separate document:

 Table of Evidence Levels of Individual Studies by Domain, Study Design, & Quality (abbreviated table below)
 Grading a Body of Evidence to Answer a Clinical Question
 Judging the Strength of a Recommendation (abbreviated table below)

Table of Evidence Levels (see note above)

Quality level Definition
Systematic review, meta-analysis, or meta-
1a† or 1b†
synthesis of multiple studies
2a or 2b Best study design for domain
3a or 3b Fair study design for domain
4a or 4b Weak study design for domain
Other: General review, expert opinion, case
5
report, consensus report, or guideline
†a = good quality study; b = lesser quality study

Table of Recommendation Strength (see note above)

Strength Definition
“Strongly recommended” There is consensus that benefits clearly outweigh risks and burdens
(or visa-versa for negative recommendations).
“Recommended” There is consensus that benefits are closely balanced with risks and burdens.
No recommendation made There is lack of consensus to direct development of a recommendation.

Dimensions: In determining the strength of a recommendation, the development group makes a considered judgment in a consensus process
that incorporates critically appraised evidence, clinical experience, and other dimensions as listed below.
1. Grade of the Body of Evidence (see note above)
2. Safety / Harm
3. Health benefit to patient (direct benefit)
4. Burden to patient of adherence to recommendation (cost, hassle, discomfort, pain, motivation, ability to adhere, time)
5. Cost-effectiveness to healthcare system (balance of cost / savings of resources, staff time, and supplies based on published studies or
onsite analysis)
6. Directness (the extent to which the body of evidence directly answers the clinical question [population/problem, intervention,
comparison, outcome])
7. Impact on morbidity/mortality or quality of life

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 4 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Appendix 1 – Likelihood ratios (LRs) for auscultation of left upper quadrant (LUQ) sound*
Study Age range LR+ (95% CIs†) LR– (95% CIs†)
Ellett 1999 6 days – 13 yrs 13.1 (1.94, 88) 0.25 (0.14, 0.43)
Neumann 1995 Adults 1.049** 0.27**
* = See Appendix 9 for definition and rule-of-thumb for likelihood ratios
† = 95% CI: 95% Confidence Interval expresses the uncertainty (precision) of a measured value; it is the range of values within which we can
be 95% sure that the true value lies. A study with a larger sample size will generate more precise measurements, resulting in a narrower
confidence interval.
**=Unable to calculate CIs from data

Appendix 2 – Likelihood ratios (LR) for aspirate pH <6 versus pH <5 to confirm gastric placement*
LR+ to rule in LR– to rule out LR+ to rule in LR– to rule out
gastric placement gastri gastric placement gastri
pH <6 c pH <5 c
Study Age range (95% CIs†) place (95% CIs†) place
ment ment
pH <6 pH <5
(95% CIs†) (95% CIs†)
24.5 0.33 Unable to calculate Unable to calculate
Metheny 1999A 14 yrs – Adult
(11.8, 51.2) (0.27, 0.40) from data from data
10.0 0.42 15.5 0.64
Phang 2004 Adults
(4.20, 23.8)‡ (0.32, 0.55) ‡ (3.84, 62.7) (0.54, 0.76)
9.55 0.64 Unable to calculate Unable to calculate
Metheny 2002 18 yrs – Adult
(1.36, 67.1) (0.52, 0.80) from data from data
5.88 0.15 12.99 0.38
Metheny 1994A 18 yrs – Adult
(4.64, 7.45) ‡ (0.12, 0.20) ‡ (8.33, 20.3) (0.34, 0.44)
5.43 0.26 Unable to calculate Unable to calculate
Westhus 2004 Birth – 14 yrs
(0.88, 33.5) (0.14, 0.48) from data from data
Neonates 4.96 0.21 2.60 0.68
Metheny 1999B
(mGA 33wk) (0.39, 62.3) (0.11, 0.41) (0.20, 33) (0.40, 1.17)
1.41 0.75 Unable to calculate Unable to calculate
Ellett 2005 3 days – 7 yrs
(0.77, 2.58) (0.40, 1.39) from data from data

Neumann 1995 Adults Unable to calculate Unable to calculate 8.30§ 0.88§

from data from data
Unable to calculate Unable to calculate 1.01 0.97
Ellett 1999 6 days – 13 yrs
from data from data (0.71, 1.45) (0.40, 2.34)
* = See Appendix 9 for definition and rule of thumb for likelihood ratios
† = 95% CI: 95% Confidence Interval expresses the uncertainty (precision) of a measured value; it is the range of values within which we can be 95% sure that
the true value lies. A study with a larger sample size will generate more precise measurements, resulting in a narrower confidence interval.
‡ = pH ≤ 6
§ = Unable to calculate CIs from data

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 5 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Appendix 3 – Mean concentrations for gastric aspirate pH, intestinal aspirate pH and respiratory samples with standard
deviation (SD)

Gastric aspirate pH Intestinal aspirate pH Respiratory aspirates/ samples pH

Study Age range mean ± SD mean ± SD mean ± SD
(Range of pH) (Range of pH) (Range of pH)

Ellett 2005 3days-7yrs 4.55 ± 0.16 Not reported No data

(1.4 - 8.5)
Westhus 2004 Birth-14yrs 4.1 ± 0.32 7.5 ± 0.33 No data
(1.2 - 8.3) (5.9 - 8.2)
Phang 2004 Adults 4.8- ± 2.3 7.1 ± 1 No data
Metheny 2002 18 yrs-adult 5.7 ± 0.1² 6.6 ± 0.1² No data

Metheny 1999B Neonates 4.32 ± 1.2 7.8ⁿ No data

(1.24 - 8.85) (7.39 - 8.20)
Metheny 1999A 14yrs-adult 3.90 ± 0.15 7.35 ± 0.06 7.73 ± 0.04
(0.87 - 8.97) (1.7 - 8.79) (5.99 - 9.07)
Metheny 1994A 18yrs-adult 3.52 ± 2.02 7.05 ± 1.26 7.38 ± 0.59
(6.74 - 8.36)
n = No SD, only 2 aspirates
2 = Standard error of the mean

Appendix 4 – Visual characteristics for gastric aspirates

Gastric Aspirate Intestinal Aspirate Respiratory Aspirate
Study Age range Colors
Colors Colors
Westhus 2004 Birth – 14 yrs  Green  Yellow No data
 Off-white  Bile-stained
 Brown
 Clear
 Colorless
Phang 2004 Adults  White  Clear yellow No data
 Frothy  Dark amber
 Sputum-like appearance
Metheny 2002 18 yrs – Adult  Off-white  Bile-stained No data
 Tan
Metheny 1999B Neonates  Off-white  Yellow/bile No data
 Tan
 Green
 Clear
 Colorless
 Bloody/brown
Metheny 1994A 18 yrs – Adult  Green  Golden/yellow No data
 Off-white  Clear
 Brown
 Colorless
 Cloudy
Metheny 1994B 18 yrs – Adult  Green (cloudy)  Yellow or bile-  Light yellow (clear or
 Off-white or tan (cloudy stained (usually blood-tinged)
 Bloody or brown clear but sometimes  Mucus (opaque)
(cloudy) cloudy)  Off-white or tan ***
 Colorless (clear)
*** = 2 samples from respiratory tract, remaining collected from thoracocentesis and suction

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 6 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Appendix 5 – Likelihood ratios (LRs) for aspirate bilirubin and CO2 to confirm gastric placement*

LR (+) aspirate LR (-) aspirate LR (+) CO2 LR (-)CO2

Study Age range bilirubin bilirubin (95% CIs1a) (95% CIs1a)
1a 1a
(95% CIs ) (95% CIs )
Metheny 2002 18 yrs – Adult 23.1 0.17 Not reported Not reported
(5.86, 91) (0.07, 0.41)
Ellett 2005 3 days – 7 yrs 4.07 0.87 1.99 0.14
(0.25, 67.3) (0.52, 1.45) (0.28, 14.1)** (0.0, 0.41)
Metheny 1999A 14 yrs- Adult 2.88 0.12 Not reported Not reported
(2.4, 3.47) (0.08, 0.19)
* = See Appendix 9 for definition and rule of thumb for likelihood ratios
1a= 95% CI: 95% Confidence Interval expresses the uncertainty (precision) of a measured value; it is the range of values within which we can be 95%
sure that the true value lies. A study with a larger sample size will generate more precise measurements, resulting in a narrower confidence interval.
** = No tubes placed in respiratory tract

Appendix 6 – Mean pepsin and trypsin concentrations for gastric and intestinal aspirates

Mean pepsin in Mean pepsin in Mean trypsin in Mean trypsin in

Study Age range gastric aspirate intestinal aspirate gastric aspirate intestinal aspirate
mean ± SD mean ± SD mean ± SD mean ± SD
Westhus 2004 Birth – 14 yrs 215.4 ± 32 ug/ml§ 24.8 ± 15.9 ug/ml§ 10.6 ± 2.9 ug/ml~ 70.4 ± 17.2 ug/ml~
Ellett 2005 3 days – 7 yrs 188 ± 23.5 ug/ml~ 38.5 ± 14.6 ug/ml~ 15.4 ± 3.8 ug/ml 181.10 ± 29.8 ug/ml~
Metheny 1999A 14 yrs- Adult 60.4 ± 6.3 ug/ml§ 4.5 ug/ml§ 6.8 ± 1.4 ug/ml 218.6 ug/ml
§ = Significant differences in pepsin & trypsin mean concentrations (p<.03)
~ = Significant differences in pepsin & trypsin mean concentrations (p<.001)

Appendix 7 – Likelihood ratios (LRs) for aspirate color and combined aspirate color and pH to confirm gastric
placement*

LR (+) aspirate LR (-) aspirate LR (+) aspirate LR (-) aspirate

Study Age range color color color and pH color and pH
(95% CIs1a) (95% CIs1a) (95% CIs1a) (95% CIs1a)
Metheny 2002 18 yrs - Adult 7.7 0.48 Data not available Data not available
(2.82, 21.1) (0.3, 0.74)
Metheny 1999B Neonates 4.08 0.38 Data not available Data not available
(0.32, 51.4) (0.21, 0.69)
Westhus 2004 Birth – 14 yrs 3.0 0.23 66.7 0.31
(0.91, 9.51) (0.11, 0.50) (4.20, 1059) (0.19, 0.49)
Metheny 1993 18 yrs – Adult 1.16 0.72 Data not available Data not available
(0.8, 1.69) 0.35, 1.49)
* = See Appendix 9 for definition and rule of thumb for likelihood ratios
1a= 95% CI: 95% Confidence Interval expresses the uncertainty (precision) of a measured value; it is the range of values within which we can be 95%
sure that the true value lies. A study with a larger sample size will generate more precise measurements, resulting in a narrower confidence interval.

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 7 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Appendix 8 :Algorithm

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 8 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Appendix 9 – Likelihood Ratio (LR)

A. Definition Figure: Likelihood Ratio Nomogram
When a health care provider evaluates a patient, he/she
determines their own “best guess” for how likely a disease is
present (or not present) at that time.

This “best guess” is dependent on:

• the disease prevalence in the community,
• the patient’s underlying medical status and current
presentation, and
• the health care provider’s experience and knowledge of
the literature.
This best guess is actually the pre-test probability.
What health care providers are looking for is a test which will
increase (or decrease) the likelihood of disease in that
patient, thus allowing them to decide to treat, not treat, or
pursue further diagnostic work up. This change in “best
guess” after diagnostic testing results in the post-test
probability of disease.
Finding the post-test probability is done easily with
likelihood ratios and a nomogram (see Figure: Likelihood
Ratio Nomogram).
What the likelihood ratio nomogram actually does is
change pre-test probability to pre-test odds, multiply by
the LR to get post-test odds, then change post-test odds
back to post-test probability…all without having to
manually perform complex calculations.
Fortunately, likelihood ratios (LRs) are determined by
inherent properties of the diagnostic test, not the prevalence
of disease in the population. By knowing the likelihood
ratio(s) for any or all given tests, health care providers can
determine which test is most informative and appropriate to
use.
B Rule of thumb
An LR value
• Greater than 10 is very helpful in increasing
diagnostic certainty
A positive result is 10 times more likely to
confirm the NGT is in the gastric area as outside
the gastric area
• Of 1 is not helpful
A positive result is just as likely to confirm the
NGT is in the gastric area as outside gastric area
• Less than 0.2 is very helpful in ruling out the
condition
A positive result is one-fifth as likely to confirm
the NGT is in the gastric area as outside the
gastric area
For more information on Likeilhood Ratios (LRs), see the
Evidence-based Decision Making Glossary on the website:
http://groups/ce/NewEBC/EBCMain.htm

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 9 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

Supporting information
Background Information
Error rates for placement of enteral tubes in any location other than the intended location show rates of 20.9% and 43.5% respectively in
pediatric settings (Ellett 2005). A small percentage of enteral tubes, reported as 1%-4% in adult intensive care settings but unknown in
pediatrics, are incorrectly placed within the respiratory tract with potentially serious consequences (Ellett 2005, Metheny 1999, Metheny 1994,
Harris & Huseby 1989). Children who are comatose, semicomatose, inactive, had swallowing problems or Argyle tubes have higher placement
errors outside the intended location (Ellett 1999) and ought to be considered higher risk for incorrect placement. Radiography is considered the
gold standard for documenting tube placement (Ellett 1999). However, routine radiological verification of pediatric and adolescent patients
increases the risk of excessive radiation exposure, increases patient and healthcare costs and slows the delivery of clinical care (Ellett 1999,
Neumann 1995). Due to these patient and healthcare risks, the evidence for the best methods to accurately verify NG placement was reviewed.
The methods reviewed include auscultation, aspirate pH, aspirate bilirubin, aspirate trypsin, aspirate pepsin, CO2 monitoring and visual
inspection of aspirates. Age-related height based methods and morphological measurements were also evaluated for accuracy in predicting NG
tube length in children.

Group/team members
Development Group
Kim Klotz, BSN, RN, Vascular Access Team, Chair
Debby Mason, MSN, CNP, RN, Advanced Practice Nursing
Lois Siegle, BSN, RN, Home Care Services
Anne Longo, MBA, BSN, RN-BC, Center for Professional Excellence, Education
Mary Porter, BSN, RN, Pediatric Intensive Care Unit
Jackie Wessel, MEd, RD, Nutrition
Amy Sapsford, RD, Nutrition
Rene’ Shelton, RN, Regional Center for Neonatal Intensive Care
Karen Burkett, MS, CNP, RN, Center for Professional Excellence, Research & EBP
Interdisciplinary team members
Sam Kocoshis, MD, Gastroenterology
Mike Farrell, MD, Chief of Staff
Rich Brilli, MD, Pediatric Intensive Care Unit
Beth Haberman, MD, Regional Center for Neonatal Intensive Care
Paul Steele, MD, Pathology and Clinical Lab
Linda Anderson, MT(ASCP), Clinical Lab

Search strategy
1. Original Search
OVID Databases
Medline, CINAHL, PubMed and the Cochrane Database for Systematic Reviews (CDSR)
OVID Filters
Publication Date 1996 to present
Limits Humans and English Language
Study Type highest quality evidence
Search Terms and MeSH Terms
Patients/Population children requiring NG tube placement, including neonatal, pediatric and adolescent patients
Intervention/Exposure aspirate, auscultation and radiology methods, NG tube length prediction methods, morphological methods, age-
related height based methods
Outcomes accurate NG tube placement, NG tube length predictions
2. Additional articles identified from reference lists and clinicians.

Applicability issues
Outcomes that are planned to be measured include:
1. Percent of patients requiring NG tubes in which placement is confirmed by aspirate and auscultation methods.
2. Percent reduction in required number of x-rays to confirm NG tube placement.
3. Percent increase in clinical staff trained in pH testing at the bedside.
4. Percent increase in clinical staff accurately measuring height/length.
5. Percent increase in Nutrition consultation for patients with NG tubes.
6. Implementation of algorithm using multiple verification methods for NG tube placement.

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 10 of 11
 Patient Services/ NGT Placement Confirmation/ BESt 024

7. Implementation of algorithm using height regression equations for tube length predictions.
8. Percent of staff correctly placing NG tubes per recommendations.

Copies of this Best Evidence Statement (BESt) are available online and may be distributed by any organization for the global purpose of
improving child health outcomes. Website address: http://www.cincinnatichildrens.org/svc/alpha/h/health-policy/ev-based/default.htm
Examples of approved uses of the BESt include the following:
• copies may be provided to anyone involved in the organization’s process for developing and implementing evidence based care;
• hyperlinks to the CCHMC website may be placed on the organization’s website;
• the BESt may be adopted or adapted for use within the organization, provided that CCHMC receives appropriate attribution on all written or
electronic documents; and
• copies may be provided to patients and the clinicians who manage their care.
Notification of CCHMC at HPCEInfo@cchmc.org for any BESt adopted, adapted, implemented or hyperlinked by the organization is
appreciated.
Additionally for more information about CCHMC Best Evidence Statements and the development process, contact the Center for Professional
Excellence/Research and Evidence-based Practice office at CPE-EBP-Group@chmcc.org.

Note
This Best Evidence Statement addresses only key points of care for the target population; it is not intended to be a comprehensive
practice guideline. These recommendations result from review of literature and practices current at the time of their formulation. This
Best Evidence Statement does not preclude using care modalities proven efficacious in studies published subsequent to the current
revision of this document. This document is not intended to impose standards of care preventing selective variances from the
recommendations to meet the specific and unique requirements of individual patients. Adherence to this Statement is voluntary. The
clinician in light of the individual circumstances presented by the patient must make the ultimate judgment regarding the priority of
any specific procedure.

Reviewed by Clinical Effectiveness

Copyright © 2009 Cincinnati Children's Hospital Medical Center; all rights reserved. Page 11 of 11