- Email: [email protected]
Pediatric orogastric and nasogastric tubes: A new formula evaluated
PEDIATRICS/ORIGINAL CONTRIBUTION
Pediatric Orogastric and Nasogastric Tubes: A New Formula Evaluated
From the Department of Pediatrics, Division of Pediatric Medicine, University of Alabama at Birmingham and The Children’s Hospital of Alabama, Birmingham, AL*; the Department of Community Health, St. Louis University School of Public Health, St. Louis, MO‡; and the Missouri Regional Poison Center and Divisions of Pediatric Emergency Medicine and Toxicology, St. Louis University and Cardinal Glennon Children’s Hospital, St. Louis, MO.§ Author contributions are provided at the end of this article. Received for publication October 27, 1999. Revisions received June 14, 2000; December 1, 2000; February 16, 2001; July 2, 2001; and July 18, 2001. Accepted for publication August 28, 2001. Presented in part at the North American Congress of Clinical Toxicology, La Jolla, CA, September 1999; and the American Academy of Pediatrics, Section of Emergency Medicine, Washington, DC, October 1999. Address for reprints: Ann E. Klasner, MD, MPH, 1600 7th Avenue South, Midtown Center, Suite 205, Birmingham, AL 35233; 205-934-2116, fax 205-975-4623; E-mail [email protected]. Copyright © 2002 by the American College of Emergency Physicians. 0196-0644/2002/$35.00 + 0 47/1/120124 doi:10.1067/mem.2002.120124
Ann E. Klasner, MD, MPH* Douglas A. Luke, PhD‡ Anthony J. Scalzo, MD§
Study objective: We sought to compare the traditional method of determining depth of gastric tube insertion, by measuring from the external landmarks of the nose or mouth, to the earlobe, to the xiphoid process (NEX method), with a graph for determining depth of gastric tube insertion that is based on patient height (graphic method). Methods: A prospective, randomized, double-blinded study comparing NEX and graphic methods for gastric tube depth of insertion was undertaken. This study included a convenience sample of pediatric emergency department patients in need of gastric intubation. Patients were block randomized, and their gastric tubes were placed to the depth derived from the particular method employed. Alternate depth of insertion was measured on all patients. Abdominal radiographs were used to determine the distance that the end of the tube was from the center of the stomach. Results: Forty-four patients each were in the NEX and graphic groups. The mean distance from the center of the stomach was –1.12 cm (SD 1.36) for the graphic group, compared with 1.31 cm (SD 3.39) for the NEX method. The difference between the 2 methods was 2.43 cm (95% confidence interval [CI] 1.33 to 3.54). Using absolute values, the mean distance from the center of the stomach was 1.26 cm (SD 1.23) for the graphic group compared with 2.60 cm (SD 2.51) for the NEX method. Using these values, the difference between the groups is 1.34 cm (95% CI 0.50 to 2.18). Conclusion: When compared with the NEX method, the graphic method demonstrates a significant ability to more consistently and accurately determine the depth of pediatric gastric tube insertion. [Klasner AE, Luke DA, Scalzo AJ. Pediatric orogastric and nasogastric tubes: a new formula evaluated. Ann Emerg Med. March 2002;39:268-272.]
2 6 8
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
INTRODUCTION
Gastric tubes (oral and nasal) are needed in the pediatric setting for a number of reasons, including decompression of bowel obstructions; contrast placement in traumatized, incoherent, uncooperative, or very young patients; and lavage and/or instillation of activated charcoal after acute ingestions from toxic substances. The placement procedure for gastric tubes, known as the NEX method, has traditionally been based on the estimation from the nose or mouth to the earlobe, to a point midway between the xiphoid process and umbilicus, with gastric auscultation for confirmation.1 This method for determining depth of gastric tube placement had been used for many years without challenge. In 1992, Scalzo et al,2 after review of 36 pediatric patients needing gastric tubes, reported a 50% (7/14) malposition rate according to a radiograph after the NEX method had been used to estimate tube depth. Scalzo et al suggested that a tube insertion formula derived from Strobel et al’s3 previously published formula for esophageal pH probe placement, which was determined on the basis of the patient’s height, might be a more consistent method of determining tube depth of insertion.2 Strobel et al’s original formula included lengths for both oral and nasal insertion as follows: Nasal Insertion: Esophageal Length (cm) = 6.7 + [0.226 × Height (cm)] Oral Insertion: Esophageal Length (cm) = 5.0 + [0.252 × Height (cm)] Scalzo et al2 modified the formula, using a correction factor to account for the confines of the more distal insertion of tubes into the stomach. This modified formula, determined on the basis of height, was displayed in graphic form and is known as the graphic method (Figure 1). Scalzo et al proceeded to use this formula/graph on 6 pediatric patients and demonstrated correct placement in 100% of these patients, based on follow-up radiographs. This study served to evaluate the accuracy of depth of gastric tube insertion using this graph. We conducted a double-blinded, randomized, controlled trial to test the hypothesis that the graphic method resulted in comparable or better accuracy in depth of tube determination compared with the traditional NEX method. M AT E R I A L S A N D M E T H O D S
This study was a prospective, double-blinded, randomized clinical trial that enrolled a convenience sample of 89
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
children who presented to a university-affiliated children’s hospital from May 1, 1996, to May 31, 1998. The study design and patient consent procedures were approved by the institutional review board. Children were eligible if they were between the ages of 6 months and 18 years and in need of a gastric intubation in the emergency department. Informed consent was obtained from the parent or guardian and assent was obtained from children who were 7 years of age or older. Children were excluded if they had prior esophageal surgery or known congenital abnormalities of the esophagus. Children were also excluded if they required emergency placement of the gastric tube. These patients included some critically ill trauma and ingestion patients, in cases in which obtaining consent might have delayed treatment. This last exclusion criterion was the primary reason for convenience sampling. Once patients were enrolled and consent was obtained, the patients had their height measured and were block randomized on the basis of their percentile height to have their gastric tube depth of insertion determined by the NEX or graphic method. Block randomization groups were formed as tall (>75%), medium (25% to 75%), and short (<25%) on the basis of standard growth curves. The nurse was unaware of which technique was selected. The nurse then measured how far down she would place the gastric tube if the NEX method was used. After the nurse completed and recorded the measurement, the principal investigator informed the nurse
Figure 1.
Estimated esophageal length for gastric tube insertion. Graph of formulas, determined on the basis of height, to determine depth of gastric tube insertion by the graphic method. OG, Orogastric; NG, nasogastric. Tube insertion depth (cm) 70 65 60 55 50 45 40 35 30 25 20
OG NG
60
70
80
90 100 110 120 130 140 150 160 170 180 190 200 Height (cm)
2 6 9
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
whether to use this measurement or the measurement that was derived and recorded from the graphic method on the basis of the patient’s height (Figure 1). The physician treating the patient specified the type of tube, size of tube, and nasogastric versus orogastric. The tube was placed and secured. The nurse noted whether she felt placement was correct by auscultation, instillation of air, and aspiration for gastric contents. A radiograph of the abdomen was obtained to verify the tube position. For female patients of appropriate age, a pregnancy test was obtained before all radiographs were taken. Two of the study investigators, who are board-certified pediatric emergency physicians and are skilled in the interpretation of pediatric radiographs, performed interpretation of the position of the gastric tubes. To ensure blinding, radiographs were held and batched into 3 separate readings. The radiographs were read in a random order without knowledge of the patient’s identity or the method used. Each radiograph was evaluated for the following questions. Is the gastric tube within the confines of the stomach? (Yes or No.) How far from the center of the stomach is the tube positioned? (This is evaluated in centimeters, with 0=center of stomach, a positive sign for tube placed too deep, and a negative sign for tube placed too shallow. The center of the stomach was defined as the midpoint between the lower esophageal sphincter or diaphragm and pylorus.) If the distance of the tube from the center of the stomach could not be determined, then the distance was determined to a depth sufficient that the most proximal hole of the gastric tube was below the diaphragm. The most proximal hole of the gastric tubes was located 10 cm from the distal tip of the gastric tube. Two analytic approaches were used to assess the differences in insertion depths of the NEX and graphic methods. First, the raw insertion measurements (in centimeters) were used to compare on average how far overinserted or underinserted the tubes were for each method. A measurement of 0 indicated perfect placement, a positive measurement indicated overinsertion, and a negative measurement indicated underinsertion. The second approach was to examine the absolute value of the insertion measurements rather than the raw value. This would provide an assessment of insertion accuracy (ie, distance from center of stomach) without specifying whether a tube was overinserted or underinserted. In both cases, Student’s t test was used to identify the difference between depth of insertion for the 2 methods. The statistical package SPSS/PC for Windows (version 10.0; SPSS Inc., Chicago, IL) was used for all statistical analyses.
2 7 0
R E S U LT S
Eighty-nine patients were prospectively enrolled. Fortyfour patients were randomly assigned to the NEX method, and 45 patients were assigned to the graphic method. One patient in the graphic group was excluded because the radiographs were unreadable as a result of poor penetration, making measurements unobtainable. This left 88 patients for analysis, with 44 in the NEX group and 44 in the graphic group. The demographics for the 2 groups, including age, height, weight, percentile height for age, percentile weight for age, sex, whether they required an orogastric or nasogastric, reason for the tube placement, and the type of tube are presented in the Table. There were no complications in either group. Using the raw insertion measurements, the mean distance from the center of the stomach was –1.12 cm (SD 1.36) for the graphic method and 1.31 cm (SD 3.39) for the NEX method. The difference between the 2 methods was 2.43 cm (95% confidence interval [CI] 1.33 to 3.54).
Table.
Patient demographics and tube placement characteristics.
Characteristic Sex Male Female Age, mo (SD) Nasogastric versus orogastric Nasogastric Orogastric Reason for gastric intubation Trauma Ingestion Obstruction Feeding Other Type of tube Tum-E-Vac* Feeding Argyle Salem Sumps† Other Missing information Height, cm (SD) Percentile height for age, % (SD) Weight, kg (SD) Percentile weight for age, % (SD)
Graphic Method (n=44)
NEX Method (n=44)
26 19 57.5 (55.3)
28 16 51.0 (53)
23 22
23 21
3 31 3 4 4
5 34 1 1 3
17 10 10 0 7 102.2 (34.4) 52.4 (34) 54.4 (19.6) 52.4 (35.7)
17 10 10 0 12 99.4 (28.1) 48 (32) 61.7 (21.9) 59 (31.6)
*
Ethox Corporation Medical Products, Buffalo, NY. Argyle Salem Sumps, St. Louis, MO.
†
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
Figure 2 displays box plots showing the distances off from the center of the stomach. Using the absolute value of the insertion method, the mean absolute value of insertion distance from the center of the stomach was 1.26 cm (SD 1.23) for the graphic method and 2.60 cm (SD 2.51) for the NEX method. The difference between the 2 methods was 1.34 cm (95% CI 0.50 to 2.18). Three tubes were read as outside of the confines of the stomach, all 3 of these tubes were in the NEX group, and all were placed distal to the confines of the stomach. Auscultation after tube placement reported correct placement in 100% of the patients. DISCUSSION
This study illustrates the dramatic variability that can occur with gastric tube placement. The NEX method demonstrated a greater mean distance off from the center of the stomach, with nearly twice as much variability compared with the graphic method (1.31 cm [SD 3.39] versus –1.12 cm [SD 1.36]). Examination of the box plots in Figure 2 reveals an important characteristic of tube insertion using the graphic method. Both the NEX and the graphic methods result in tube placement that on average is close to the center of the stomach. However, Figure 2 makes it clear
Figure 2.
Comparison of distances from center of the stomach by tube placement method. Solid line, Median distance from center of stomach; dotted line, mean distance from center of stomach.
Distance from stomach center (cm) 8 6 4 2 0 –2 –4 NEX method
MARCH 2002
39:3
Graphic method
ANNALS OF EMERGENCY MEDICINE
that the variability of placement is much greater for the traditional NEX method than for the graphic method. In terms of underestimation, both groups were similar, with the worst underinsertion as –5 cm. However, in terms of overinsertion, the NEX method seemed less accurate. For example, tubes were overinserted by 4 cm or more in 25% of the NEX cases. In comparison, the tubes were not overinserted by more than 2 cm in any of the graphic cases. The graphic method led to placement nearer to the center of the stomach. In addition, the variability was dramatically reduced when the graphic method was used. The graphic method was more accurate, demonstrated decreased variability, and therefore allowed gastric tube placement closer to a designated target of the center of the stomach. We speculate that this will allow for more reliable placement of gastric tubes and fewer problems. Better accuracy, increased consistency, and decreased variability when attempting to place a gastric tube in the designated spot of the center of the stomach were displayed when the graphic method of tube placement was used. Correct tube placement was confirmed in 100% of the cases by auscultation performed by nurses after tube placement and before radiographs. After reviewing radiographs, 3 (3.4%) of the gastric tubes placed were noted outside of the confines of the stomach. All of these tubes (3/44) were from the NEX group. We do not expect nurses to be able to tell where in the confines of the stomach the gastric tube might be located, but we would anticipate that those tubes outside of the stomach would be identified. Although this information is limited by numbers, it does suggest a trend. We can speculate that perhaps auscultation is an ineffective method to determine whether gastric tubes are within the confines of the stomach. Our study has the primary limitation of being a convenience sampling of patients. We excluded patients who were in need of rapid gastric tube placement, in cases in which the gastric tube portion of the treatment plan was so critical that proper enrollment in the study was precluded. This primarily represented critical trauma patients. This convenience sampling is a potential cause of selection bias. Despite the fact that this potential for selection bias is present, we believe that it does not represent a significant number of patients and thereby would have resulted in little alteration in the outcome of the study. To decrease this bias in future testing of this method, the formula could be incorporated into the Broselow tape method. It has been well demonstrated when using the Broselow tape that there is a correlation between patient height and determination of weight and drug doses in the pediatric population.4 Using this infor-
2 7 1
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
mation, our formula, which is also based on patient height, could be incorporated into the Broselow tape. This could allow for easier use during trauma or code situations. Our study is also fraught with difficulty in determining the accuracy and consistency of the definitions proposed. It is questionable whether one can truly estimate the center of the stomach on plain radiographic screening alone. A more ideal situation would include a barium swallow or esophagoscopy. In lieu of these techniques, we offer the best alternative. We realize the limitations of plain radiograph interpretation and attempt to account for it by using the same readers to interpret all radiographs for consistency. We had the radiographs read in batches to increase consistency in interpretation, and we kept the readers blinded to the method used on each radiograph. These provisions helped to make the plain radiograph interpretation as consistent as possible. Gastric tube insertion is not a benign procedure. Complications have been described in the medical literature. Some of these complications include activated charcoal instillation into the lung, gastric tract perforation with charcoal peritoneum, impacted orogastric tubes in the esophagus, transbronchial intubation of the right pleural space, acute mediastinal widening from esophageal perforation, and other cases of esophageal perforation and tears.5-11 Presumably, each of these gastric tubes were placed using the traditional NEX technique. In this study, we have not addressed the usability or clinical significance of tube placement. However, we did demonstrate that, given a specific target, the graphic method was more accurate, consistent, and reliable. We can speculate that these findings might decrease the potential rate of complications. In addition, this method of tube insertion is noninvasive and quick. Given an option of techniques (NEX versus graphic), the authors feel that choosing a method with proven accuracy is better. We conclude that the graphic method of determining depth of gastric tube insertion, determined on the basis of patient height, is more accurate and consistent than the NEX method.
3. Strobel CT, Byrne WJ, Ament ME, et al. Correlation of esophageal lengths in children with height: application to the Tuttle test without prior esophageal manometry. J Pediatr. 1979;94:81-84. 4. Lubitz DS, Seidel JS, Chameides L, et al. A rapid method for estimating weight and resuscitation drug dosages from length in the pediatric age group. Ann Emerg Med. 1988;17:576-581. 5. Harris CR, Filandrinos D. Accidental administration of activated charcoal into the lung: aspiration by proxy. Ann Emerg Med. 1992;22:1470-1473. 6. Mariani PJ, Pook N. Gastrointestinal tract perforation with charcoal: peritoneum complicating orogastric intubation and lavage. Ann Emerg Med. 1993;22:606-609. 7. Thoma ME, Glauser JM. Use of glucagon for removal of an orogastric lavage tube. Am J Emerg Med. 1995;13:219-222. 8. El-Gamel A, Watson DCT. Transbronchial intubation of the right pleural space: a rare complication of nasogastric intubation with a polyvinylchloride tube-a case study. Heart Lung. 1993;22:224-225. 9. Padmanbhan K, Gadde H, Vora S, et al. Acute mediastinal widening following endotracheal intubation and gastric lavage. West J Med. 1991;155:419-420. 10. Askenasi R, Abramowicz M, Jeanmart J, et al. Esophageal perforation: an unusual complication of gastric lavage [abstract]. Ann Emerg Med. 1984;13:146. 11. Wald P, Stern J, Weiner B, et al. Esophageal tear following forceful removal of an impacted oral-gastric lavage tube. Ann Emerg Med. 1986;15:135-137.
Author contributions: AEK and AJS conceived the study and designed the trial. AEK acquired the data. DAL provided statistical advice on study design and analyzed the data. AEK drafted the manuscript, and all authors contributed substantially to its revision. AEK takes responsibility for the paper as a whole.
REFERENCES 1. Foster R, Hunsberger MM, Anderson J. Family-Centered Nursing Care of Children. Philadelphia, PA: WB Saunders; 1989:1388-1390, 2063. 2. Scalzo AJ, Tominack RL, Thompson MW. Malposition of pediatric gastric lavage tubes demonstrated radiographically. J Emerg Med. 1992;10:581-586.
2 7 2
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
Pediatric Orogastric and Nasogastric Tubes: A New Formula Evaluated
From the Department of Pediatrics, Division of Pediatric Medicine, University of Alabama at Birmingham and The Children’s Hospital of Alabama, Birmingham, AL*; the Department of Community Health, St. Louis University School of Public Health, St. Louis, MO‡; and the Missouri Regional Poison Center and Divisions of Pediatric Emergency Medicine and Toxicology, St. Louis University and Cardinal Glennon Children’s Hospital, St. Louis, MO.§ Author contributions are provided at the end of this article. Received for publication October 27, 1999. Revisions received June 14, 2000; December 1, 2000; February 16, 2001; July 2, 2001; and July 18, 2001. Accepted for publication August 28, 2001. Presented in part at the North American Congress of Clinical Toxicology, La Jolla, CA, September 1999; and the American Academy of Pediatrics, Section of Emergency Medicine, Washington, DC, October 1999. Address for reprints: Ann E. Klasner, MD, MPH, 1600 7th Avenue South, Midtown Center, Suite 205, Birmingham, AL 35233; 205-934-2116, fax 205-975-4623; E-mail [email protected]. Copyright © 2002 by the American College of Emergency Physicians. 0196-0644/2002/$35.00 + 0 47/1/120124 doi:10.1067/mem.2002.120124
Ann E. Klasner, MD, MPH* Douglas A. Luke, PhD‡ Anthony J. Scalzo, MD§
Study objective: We sought to compare the traditional method of determining depth of gastric tube insertion, by measuring from the external landmarks of the nose or mouth, to the earlobe, to the xiphoid process (NEX method), with a graph for determining depth of gastric tube insertion that is based on patient height (graphic method). Methods: A prospective, randomized, double-blinded study comparing NEX and graphic methods for gastric tube depth of insertion was undertaken. This study included a convenience sample of pediatric emergency department patients in need of gastric intubation. Patients were block randomized, and their gastric tubes were placed to the depth derived from the particular method employed. Alternate depth of insertion was measured on all patients. Abdominal radiographs were used to determine the distance that the end of the tube was from the center of the stomach. Results: Forty-four patients each were in the NEX and graphic groups. The mean distance from the center of the stomach was –1.12 cm (SD 1.36) for the graphic group, compared with 1.31 cm (SD 3.39) for the NEX method. The difference between the 2 methods was 2.43 cm (95% confidence interval [CI] 1.33 to 3.54). Using absolute values, the mean distance from the center of the stomach was 1.26 cm (SD 1.23) for the graphic group compared with 2.60 cm (SD 2.51) for the NEX method. Using these values, the difference between the groups is 1.34 cm (95% CI 0.50 to 2.18). Conclusion: When compared with the NEX method, the graphic method demonstrates a significant ability to more consistently and accurately determine the depth of pediatric gastric tube insertion. [Klasner AE, Luke DA, Scalzo AJ. Pediatric orogastric and nasogastric tubes: a new formula evaluated. Ann Emerg Med. March 2002;39:268-272.]
2 6 8
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
INTRODUCTION
Gastric tubes (oral and nasal) are needed in the pediatric setting for a number of reasons, including decompression of bowel obstructions; contrast placement in traumatized, incoherent, uncooperative, or very young patients; and lavage and/or instillation of activated charcoal after acute ingestions from toxic substances. The placement procedure for gastric tubes, known as the NEX method, has traditionally been based on the estimation from the nose or mouth to the earlobe, to a point midway between the xiphoid process and umbilicus, with gastric auscultation for confirmation.1 This method for determining depth of gastric tube placement had been used for many years without challenge. In 1992, Scalzo et al,2 after review of 36 pediatric patients needing gastric tubes, reported a 50% (7/14) malposition rate according to a radiograph after the NEX method had been used to estimate tube depth. Scalzo et al suggested that a tube insertion formula derived from Strobel et al’s3 previously published formula for esophageal pH probe placement, which was determined on the basis of the patient’s height, might be a more consistent method of determining tube depth of insertion.2 Strobel et al’s original formula included lengths for both oral and nasal insertion as follows: Nasal Insertion: Esophageal Length (cm) = 6.7 + [0.226 × Height (cm)] Oral Insertion: Esophageal Length (cm) = 5.0 + [0.252 × Height (cm)] Scalzo et al2 modified the formula, using a correction factor to account for the confines of the more distal insertion of tubes into the stomach. This modified formula, determined on the basis of height, was displayed in graphic form and is known as the graphic method (Figure 1). Scalzo et al proceeded to use this formula/graph on 6 pediatric patients and demonstrated correct placement in 100% of these patients, based on follow-up radiographs. This study served to evaluate the accuracy of depth of gastric tube insertion using this graph. We conducted a double-blinded, randomized, controlled trial to test the hypothesis that the graphic method resulted in comparable or better accuracy in depth of tube determination compared with the traditional NEX method. M AT E R I A L S A N D M E T H O D S
This study was a prospective, double-blinded, randomized clinical trial that enrolled a convenience sample of 89
MARCH 2002
39:3
ANNALS OF EMERGENCY MEDICINE
children who presented to a university-affiliated children’s hospital from May 1, 1996, to May 31, 1998. The study design and patient consent procedures were approved by the institutional review board. Children were eligible if they were between the ages of 6 months and 18 years and in need of a gastric intubation in the emergency department. Informed consent was obtained from the parent or guardian and assent was obtained from children who were 7 years of age or older. Children were excluded if they had prior esophageal surgery or known congenital abnormalities of the esophagus. Children were also excluded if they required emergency placement of the gastric tube. These patients included some critically ill trauma and ingestion patients, in cases in which obtaining consent might have delayed treatment. This last exclusion criterion was the primary reason for convenience sampling. Once patients were enrolled and consent was obtained, the patients had their height measured and were block randomized on the basis of their percentile height to have their gastric tube depth of insertion determined by the NEX or graphic method. Block randomization groups were formed as tall (>75%), medium (25% to 75%), and short (<25%) on the basis of standard growth curves. The nurse was unaware of which technique was selected. The nurse then measured how far down she would place the gastric tube if the NEX method was used. After the nurse completed and recorded the measurement, the principal investigator informed the nurse
Figure 1.
Estimated esophageal length for gastric tube insertion. Graph of formulas, determined on the basis of height, to determine depth of gastric tube insertion by the graphic method. OG, Orogastric; NG, nasogastric. Tube insertion depth (cm) 70 65 60 55 50 45 40 35 30 25 20
OG NG
60
70
80
90 100 110 120 130 140 150 160 170 180 190 200 Height (cm)
2 6 9
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
whether to use this measurement or the measurement that was derived and recorded from the graphic method on the basis of the patient’s height (Figure 1). The physician treating the patient specified the type of tube, size of tube, and nasogastric versus orogastric. The tube was placed and secured. The nurse noted whether she felt placement was correct by auscultation, instillation of air, and aspiration for gastric contents. A radiograph of the abdomen was obtained to verify the tube position. For female patients of appropriate age, a pregnancy test was obtained before all radiographs were taken. Two of the study investigators, who are board-certified pediatric emergency physicians and are skilled in the interpretation of pediatric radiographs, performed interpretation of the position of the gastric tubes. To ensure blinding, radiographs were held and batched into 3 separate readings. The radiographs were read in a random order without knowledge of the patient’s identity or the method used. Each radiograph was evaluated for the following questions. Is the gastric tube within the confines of the stomach? (Yes or No.) How far from the center of the stomach is the tube positioned? (This is evaluated in centimeters, with 0=center of stomach, a positive sign for tube placed too deep, and a negative sign for tube placed too shallow. The center of the stomach was defined as the midpoint between the lower esophageal sphincter or diaphragm and pylorus.) If the distance of the tube from the center of the stomach could not be determined, then the distance was determined to a depth sufficient that the most proximal hole of the gastric tube was below the diaphragm. The most proximal hole of the gastric tubes was located 10 cm from the distal tip of the gastric tube. Two analytic approaches were used to assess the differences in insertion depths of the NEX and graphic methods. First, the raw insertion measurements (in centimeters) were used to compare on average how far overinserted or underinserted the tubes were for each method. A measurement of 0 indicated perfect placement, a positive measurement indicated overinsertion, and a negative measurement indicated underinsertion. The second approach was to examine the absolute value of the insertion measurements rather than the raw value. This would provide an assessment of insertion accuracy (ie, distance from center of stomach) without specifying whether a tube was overinserted or underinserted. In both cases, Student’s t test was used to identify the difference between depth of insertion for the 2 methods. The statistical package SPSS/PC for Windows (version 10.0; SPSS Inc., Chicago, IL) was used for all statistical analyses.
2 7 0
R E S U LT S
Eighty-nine patients were prospectively enrolled. Fortyfour patients were randomly assigned to the NEX method, and 45 patients were assigned to the graphic method. One patient in the graphic group was excluded because the radiographs were unreadable as a result of poor penetration, making measurements unobtainable. This left 88 patients for analysis, with 44 in the NEX group and 44 in the graphic group. The demographics for the 2 groups, including age, height, weight, percentile height for age, percentile weight for age, sex, whether they required an orogastric or nasogastric, reason for the tube placement, and the type of tube are presented in the Table. There were no complications in either group. Using the raw insertion measurements, the mean distance from the center of the stomach was –1.12 cm (SD 1.36) for the graphic method and 1.31 cm (SD 3.39) for the NEX method. The difference between the 2 methods was 2.43 cm (95% confidence interval [CI] 1.33 to 3.54).
Table.
Patient demographics and tube placement characteristics.
Characteristic Sex Male Female Age, mo (SD) Nasogastric versus orogastric Nasogastric Orogastric Reason for gastric intubation Trauma Ingestion Obstruction Feeding Other Type of tube Tum-E-Vac* Feeding Argyle Salem Sumps† Other Missing information Height, cm (SD) Percentile height for age, % (SD) Weight, kg (SD) Percentile weight for age, % (SD)
Graphic Method (n=44)
NEX Method (n=44)
26 19 57.5 (55.3)
28 16 51.0 (53)
23 22
23 21
3 31 3 4 4
5 34 1 1 3
17 10 10 0 7 102.2 (34.4) 52.4 (34) 54.4 (19.6) 52.4 (35.7)
17 10 10 0 12 99.4 (28.1) 48 (32) 61.7 (21.9) 59 (31.6)
*
Ethox Corporation Medical Products, Buffalo, NY. Argyle Salem Sumps, St. Louis, MO.
†
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
Figure 2 displays box plots showing the distances off from the center of the stomach. Using the absolute value of the insertion method, the mean absolute value of insertion distance from the center of the stomach was 1.26 cm (SD 1.23) for the graphic method and 2.60 cm (SD 2.51) for the NEX method. The difference between the 2 methods was 1.34 cm (95% CI 0.50 to 2.18). Three tubes were read as outside of the confines of the stomach, all 3 of these tubes were in the NEX group, and all were placed distal to the confines of the stomach. Auscultation after tube placement reported correct placement in 100% of the patients. DISCUSSION
This study illustrates the dramatic variability that can occur with gastric tube placement. The NEX method demonstrated a greater mean distance off from the center of the stomach, with nearly twice as much variability compared with the graphic method (1.31 cm [SD 3.39] versus –1.12 cm [SD 1.36]). Examination of the box plots in Figure 2 reveals an important characteristic of tube insertion using the graphic method. Both the NEX and the graphic methods result in tube placement that on average is close to the center of the stomach. However, Figure 2 makes it clear
Figure 2.
Comparison of distances from center of the stomach by tube placement method. Solid line, Median distance from center of stomach; dotted line, mean distance from center of stomach.
Distance from stomach center (cm) 8 6 4 2 0 –2 –4 NEX method
MARCH 2002
39:3
Graphic method
ANNALS OF EMERGENCY MEDICINE
that the variability of placement is much greater for the traditional NEX method than for the graphic method. In terms of underestimation, both groups were similar, with the worst underinsertion as –5 cm. However, in terms of overinsertion, the NEX method seemed less accurate. For example, tubes were overinserted by 4 cm or more in 25% of the NEX cases. In comparison, the tubes were not overinserted by more than 2 cm in any of the graphic cases. The graphic method led to placement nearer to the center of the stomach. In addition, the variability was dramatically reduced when the graphic method was used. The graphic method was more accurate, demonstrated decreased variability, and therefore allowed gastric tube placement closer to a designated target of the center of the stomach. We speculate that this will allow for more reliable placement of gastric tubes and fewer problems. Better accuracy, increased consistency, and decreased variability when attempting to place a gastric tube in the designated spot of the center of the stomach were displayed when the graphic method of tube placement was used. Correct tube placement was confirmed in 100% of the cases by auscultation performed by nurses after tube placement and before radiographs. After reviewing radiographs, 3 (3.4%) of the gastric tubes placed were noted outside of the confines of the stomach. All of these tubes (3/44) were from the NEX group. We do not expect nurses to be able to tell where in the confines of the stomach the gastric tube might be located, but we would anticipate that those tubes outside of the stomach would be identified. Although this information is limited by numbers, it does suggest a trend. We can speculate that perhaps auscultation is an ineffective method to determine whether gastric tubes are within the confines of the stomach. Our study has the primary limitation of being a convenience sampling of patients. We excluded patients who were in need of rapid gastric tube placement, in cases in which the gastric tube portion of the treatment plan was so critical that proper enrollment in the study was precluded. This primarily represented critical trauma patients. This convenience sampling is a potential cause of selection bias. Despite the fact that this potential for selection bias is present, we believe that it does not represent a significant number of patients and thereby would have resulted in little alteration in the outcome of the study. To decrease this bias in future testing of this method, the formula could be incorporated into the Broselow tape method. It has been well demonstrated when using the Broselow tape that there is a correlation between patient height and determination of weight and drug doses in the pediatric population.4 Using this infor-
2 7 1
PEDIATRIC GASTRIC TUBE FORMULA Klasner, Luke & Scalzo
mation, our formula, which is also based on patient height, could be incorporated into the Broselow tape. This could allow for easier use during trauma or code situations. Our study is also fraught with difficulty in determining the accuracy and consistency of the definitions proposed. It is questionable whether one can truly estimate the center of the stomach on plain radiographic screening alone. A more ideal situation would include a barium swallow or esophagoscopy. In lieu of these techniques, we offer the best alternative. We realize the limitations of plain radiograph interpretation and attempt to account for it by using the same readers to interpret all radiographs for consistency. We had the radiographs read in batches to increase consistency in interpretation, and we kept the readers blinded to the method used on each radiograph. These provisions helped to make the plain radiograph interpretation as consistent as possible. Gastric tube insertion is not a benign procedure. Complications have been described in the medical literature. Some of these complications include activated charcoal instillation into the lung, gastric tract perforation with charcoal peritoneum, impacted orogastric tubes in the esophagus, transbronchial intubation of the right pleural space, acute mediastinal widening from esophageal perforation, and other cases of esophageal perforation and tears.5-11 Presumably, each of these gastric tubes were placed using the traditional NEX technique. In this study, we have not addressed the usability or clinical significance of tube placement. However, we did demonstrate that, given a specific target, the graphic method was more accurate, consistent, and reliable. We can speculate that these findings might decrease the potential rate of complications. In addition, this method of tube insertion is noninvasive and quick. Given an option of techniques (NEX versus graphic), the authors feel that choosing a method with proven accuracy is better. We conclude that the graphic method of determining depth of gastric tube insertion, determined on the basis of patient height, is more accurate and consistent than the NEX method.
3. Strobel CT, Byrne WJ, Ament ME, et al. Correlation of esophageal lengths in children with height: application to the Tuttle test without prior esophageal manometry. J Pediatr. 1979;94:81-84. 4. Lubitz DS, Seidel JS, Chameides L, et al. A rapid method for estimating weight and resuscitation drug dosages from length in the pediatric age group. Ann Emerg Med. 1988;17:576-581. 5. Harris CR, Filandrinos D. Accidental administration of activated charcoal into the lung: aspiration by proxy. Ann Emerg Med. 1992;22:1470-1473. 6. Mariani PJ, Pook N. Gastrointestinal tract perforation with charcoal: peritoneum complicating orogastric intubation and lavage. Ann Emerg Med. 1993;22:606-609. 7. Thoma ME, Glauser JM. Use of glucagon for removal of an orogastric lavage tube. Am J Emerg Med. 1995;13:219-222. 8. El-Gamel A, Watson DCT. Transbronchial intubation of the right pleural space: a rare complication of nasogastric intubation with a polyvinylchloride tube-a case study. Heart Lung. 1993;22:224-225. 9. Padmanbhan K, Gadde H, Vora S, et al. Acute mediastinal widening following endotracheal intubation and gastric lavage. West J Med. 1991;155:419-420. 10. Askenasi R, Abramowicz M, Jeanmart J, et al. Esophageal perforation: an unusual complication of gastric lavage [abstract]. Ann Emerg Med. 1984;13:146. 11. Wald P, Stern J, Weiner B, et al. Esophageal tear following forceful removal of an impacted oral-gastric lavage tube. Ann Emerg Med. 1986;15:135-137.
Author contributions: AEK and AJS conceived the study and designed the trial. AEK acquired the data. DAL provided statistical advice on study design and analyzed the data. AEK drafted the manuscript, and all authors contributed substantially to its revision. AEK takes responsibility for the paper as a whole.
REFERENCES 1. Foster R, Hunsberger MM, Anderson J. Family-Centered Nursing Care of Children. Philadelphia, PA: WB Saunders; 1989:1388-1390, 2063. 2. Scalzo AJ, Tominack RL, Thompson MW. Malposition of pediatric gastric lavage tubes demonstrated radiographically. J Emerg Med. 1992;10:581-586.
2 7 2
ANNALS OF EMERGENCY MEDICINE
39:3 MARCH 2002