Intraluminal esophageal pH monitoring, even though it has some limitations, is still the gold standard to measure esophageal acid exposure and to diagnose gastroesophageal reflux disease (GERD) (1). Catheter-based pH electrodes are by convention positioned 5 cm above the superior margin of the lower esophageal sphincter (LES) in adults (2), but in children, it can vary between 3 and 5 cm, which corresponds to a placement equivalent to 87% of the length of the esophagus from the nares as described by Euler and Ament (3). This localization minimizes potential artifacts that could result from catheter migration during deglutition (2). Proper placement is crucial to avoid an over- or underestimated diagnosis of GERD.
In adults, it is recommended that the pH probe be placed under manometric guidance (4,5). In children, equations have been developed to find simple, reliable ways of calculating the esophageal length (EL) to avoid the need for esophageal manometry (EM) and to facilitate the widespread use of pH probes. The Strobel formula (5 cm + 0.252 × height [cm]) is the most common and widely used mathematical equation for EL.
Despite the acceptance of these equations in pediatric clinical practice, without any prospective validity studies, there is a lack of uniformity in pH probe placement in both clinical and research settings. For placement of the pH probe some centers use a formula; others use fluoroscopy, endoscopy, or the pH step-up technique (6). To ensure accurate placement, we have decided to revisit 1 of these methods. The aim of this study was to confirm prospectively whether the most widely used equation, the Strobel formula, is accurate in calculating the EL in our pediatric population. Our secondary goal was to evaluate whether a new formula could be developed from our nonsurgical and surgical populations, determining whether there is a correlation between the patients’ height and measured EL by EM.
All conventional esophageal manometric tracings from October 2000 to October 2009 were reviewed at the Montreal Children's Hospital. Medical records were also reviewed to retrieve patients’ demographic characteristics such as age, height, medical history, and indications for the procedure. Neurologically impaired children and patients with scoliosis were excluded from the study because the height measurement may not be accurate. Patients were divided in 2 groups according to their medical history: surgical and nonsurgical. In the nonsurgical group, presurgery patients with achalasia were analyzed separately because esophageal dilatation and shortening of the esophagus both may affect the length of the esophagus. EM was performed using a transnasal 4-channel perfusion catheter with an outside diameter of 3.2 mm. For adolescent patients, the recording ports were spaced at 5-cm intervals, and for younger patients, at 3-cm intervals. The catheter was perfused with a pneumohydraulic perfusion pump (Mui Scientific, Mississauga, Ontario, Canada), and the software used to analyze the data was Gastrotrac (Alpine Biomed Corp, Orange, CA). We used a standardized method that has already been reported (7). None of the patients received sedation before catheter placement. Resting intragastric pressure was used as a zero reference for all data acquired. Resting LES pressure was measured using a slow pull-through technique (catheter withdrawn 0.5 to 1 cm at a time). For each channel, we noted the distance from the nostril to the high-pressure zone. For most of the manometry studies, 5 swallows were obtained on 2 of 4 channels for a total of 10 wet swallows per study. The location of the high-pressure zone of the LES was an average of the 2 distal channels where wet swallows were performed. The measured esophageal length obtained through the manometry study was then compared with the calculated length of the esophagus using the Strobel formula (8).
To evaluate whether a correlation exists between the patient's height and the measured EL by EM, we used a scatter diagram. We did a linear regression using SPSS software (SPSS Inc, Chicago, IL), and the equation was compared with the 2 age groups in our nonsurgical population: 3 to 10 years old (61 EM) and 11 to 18 years old (70 EM). We also looked at the same correlation in our surgical group. P < 0.05 was considered statistically significant.
Between October 2000 and October 2009, 162 children between 3 and 18 years inclusively had a total of 202 esophageal manometries. We analyzed separately the 12 presurgery patients with achalasia, and we divided the remaining 150 patients (186 EM) into 2 groups (Fig. 1). The group without previous surgery included 116 children for a total of 131 EM. Of these EM studies, 120 were done before placing a pH probe, which was indicated for GERD symptoms (75), dysphagia (23), respiratory symptoms (16), and abdominal pain (11). Other indications for the test included vomiting, failure to thrive, feeding disorders, gastroparesis, and nausea. Some patients had a repeat EM as a control to a previous abnormal study. In the group with a previous history of esophageal surgery, 55 EM studies were performed on 34 children. Among these patients, 14 patients had surgery for esophageal atresia–tracheoesophageal fistula (EA–TEF), 7 had a fundoplication, 2 had both surgery for EA–TEF and fundoplication, and 8 had a Heller myotomy. Three patients had other types of surgeries for congenital diaphragmatic hernia, rhabdomyosarcoma, and medulloblastoma.
The Strobel formula was used to calculate the EL in each patient, and it was compared with the measured EL obtained by EM (Fig. 2). The calculated EL by the Strobel formula was longer than the measured EL by an average of 3.0 ± 0.32 cm in the nonsurgical group and by 3.3 ± 0.54 cm in the surgical group (P < 0.001). A smaller difference between the measured and calculated EL was found in the presurgery achalasia group (1.2 ± 0.3 cm long), but only 12 patients were included.
A scatter diagram and regression line showing the relation of the LES location to the height for all of the patients is shown in Fig. 3. The regression equation representing this relation follows: L = 0.216 (H) + 7.13 with a coefficient of correlation (r) of 0.92 and a coefficient of determination (r 2) of 0.85. A similar correlation was found within the 2 age subgroups: L = 0.199 (H) + 9.28 [r 2 = 0.76] in the 3 years old or older to 10 years old or younger group and L = 0.230 (H) + 4.91 [r 2 = 0.7] in the older than 10 years to 18 years or younger group. All coefficients of determination from these equations were highly significant (P < 0.001).
Another correlation was found between the height and the measured EL in the surgical group. Because the type of surgery performed directly affects the EL, the regression equation varies among the surgical subgroups. For the group that had a EA–TEF repair, the linear regression equation is L = 0.223 (H) + 5.28 [r = 0.94, r 2 = 0.88], which is similar to the linear regression equation found in our nonsurgical pediatric population (Fig. 4). On the contrary, for the Heller myotomy group, the linear regression equation is L = 0.467 (H) − 32.5, but it is associated with a poor coefficient of determination [r 2 = 0.65].
Accurate placement of the catheter for pH study is essential to avoid the under- or overdiagnosis of GERD. An incorrect placement could have considerable clinical effect. This is the first study to apply prospectively the Strobel formula to calculate the esophageal length in a large cohort of children older than 3 years and to compare the results of this mathematical equation with the EM value. Our results showed that the most common and widely used formula, Strobel, could not predict with precision the LES location. On average, there was a 3-cm difference between the calculated EL and the measured EL.
The Strobel formula (8) was published in 1979 to eliminate the need for EM before the Tuttle test and to make the Tuttle test more accessible. The formula was elaborated from the data of LES location by manometry and the heights of children ages between 3 weeks and 235 months. It was based on an extremely small sample size—in only 30 of the 124 manometries performed in the study was the catheter passed through the nose. The manometric technique was also slightly different because the LES was located with a rapid pull-through instead of a slow pull-through technique.
The accuracy of the Strobel formula in the premature infant population has been studied but with conflicting results. Omari et al (9) studied the correlation between 156 premature infant heights and measured EL by EM using a methodology similar to the Strobel formula. In that study, the Strobel formula accurately predicted the distance from the nares to the LES in premature infants with a body length >40 cm, but not in shorter infants. Omari et al did not recommend the use of the Strobel formula for preterm babies shorter than 40 cm because it overestimates the EL.
Emmerson et al (10) had studied 26 preterm infants and found a positive correlation between the calculation of the pH probe position using the Strobel formula and the position on x-ray; however, the study did not examine a correlation between x-ray position and manometry, which is recognized to be the gold standard to assess the esophageal length in adults.
To predict the LES location, Staiano and Clouse (11) examined the potential relation between a patient's height and LES location by manometric determination in 123 pediatric patients. A highly significant linear correlation between body length and esophageal length was found, but precision decreases as the patient's height increases during childhood. They concluded that the LES location can be predicted from height in patients up to 2 years of age with this equation (L = 0.22 [H] + 4.92), which differs slightly from the Strobel formula.
Moreover, the European Society for Pediatric Gastroenterology and Nutrition statement in 1992 suggested that the Strobel formula may be less precise and less useful in children with height >100 cm (12). This statement is reinforced by Fig. 5, which demonstrates the difference between our linear equation and the Strobel formula. The major difference between the 2 formulas appears in children ages older than 3 years, but tends to disappear when we extrapolate our equation for younger and shorter patients, suggesting that below a height of 85 cm, both formulas seem to have similar accuracy.
As in previous studies, one of our limitations is that patients with positive or negative pH probe studies were not analyzed separately. Therefore, we cannot evaluate the real effect of the different placements, by formula or EM, on the final results of the pH probe studies and their clinical effect. We also may have underestimated the possible proximal migration of the LES, secondary to acid, as demonstrated in a previous study (13).
This is also the first study that looks at the relation between the LES position and height in a surgical population. Interestingly, in the postsurgical group of EA–TEF patients, the linear regression was almost similar to the one derived from the nonsurgical group. Because these surgeries are performed in early infancy, this may suggest that the potential growth of the esophagus is not affected, and the esophagus is able to achieve a length similar to the nonsurgical population. On the contrary, in the post-Heller myotomy group, the poor r 2 from the linear regression suggests that either the surgery arrives at a moment when linear growth is almost completed or that there is a shortening of the esophagus secondary to its dilatation; however, our sample size was extremely small to draw formal conclusions.
Finally, the result of our study confirms that a relation does exist between height and the LES position. The height of these children was found to be predictive of the LES localization (L = 0.216 [H] + 7.13) with a coefficient of correlation (r) of 0.92 and a coefficient of determination (r 2) of 0.85. Looking at the 2 sub-age groups (≥3–≤10 and >10–≤18 years old), it becomes impossible to recommend a specific and precise equation for each of them because the coefficient of determination (r 2) is found to be lower, indicating a significant decrease in the precision of these formulas. Then a global use of these formulas becomes difficult to recommend. On the contrary, EM is often considered to be invasive in children and often not accessible. Our new proposed linear equation could be useful in day-to-day pediatric practice because it is definitively more accurate than the one used in most centers based on the study by Strobel et al.
Despite the fact that high-resolution manometry (HRM) seems less invasive for children compared with conventional manometry and should be used more often in the near future, the accuracy of this technique needs to be carefully assessed to determine the location of LES. A previous study on adults showed that HRM could consistently overestimate LES length, which then may affect proper pH probe positioning (14). On the contrary, 70% of the population described in this study had hiatal hernia, which is much less frequent in the pediatric population. These findings may then not be applicable to children and adolescents. We believe that a prospective data collection on HRM LES measurements and use of formulas is needed in the future to validate the accuracy of HRM for LES assessment and to determine formula for each sub-age group.
The gold standard for placing a pH catheter in adults is by performing EM. In children, locating the LES for esophageal pH monitoring is also a clear indication to use EM (15); however, EM is perceived as invasive in children and is not widely accessible. A correlation exists between height and the LES position as shown by Strobel et al, but this formula is inaccurate in calculating the esophageal length in the pediatric population ages between 3 and 18 years. If EM is unavailable, the formula of Staiano could be considered for the placement of a pH catheter in children younger than 2 years and, as well, our new mathematical equation could be considered for children between the age of 3 and 18 years. The multiple limitations of different formulas cannot eliminate the need for EM because no equation will ever be as accurate as manometric placement. As HRM becomes more available and easier to perform, we hope that pH probe placement by esophageal HRM should become a standard clinical practice in the pediatric population once its accuracy is demonstrated for the LES measurements.
1. Pediatric Gastroesophageal Reflux Clinical Practice Guidelines: joint recommendations of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) and the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN). J Pediatr Gastroenterol Nutr
2. Hirano I, Richter JE. The Practice Parameters Committee of the American College of Gastroenterology. ACG Practice Guidelines: esophageal reflux testing. Am J Gastroenterol
3. Euler AR, Ament ME. Detection of gastroesophageal reflux in the pediatric-age patient by esophageal intraluminal pH probe measurement (Tuttle test). Pediatrics
4. Pace F, Annese V, Ceccatelli P, et al. Ambulatory esophageal pH-metry. Position Paper of the Working Team on Oesophageal pH-Metry by the GISMAD (Gruppo Italiona di Studio sulla Motilita dell’Apparato Digerente). Dig Liver Dis
5. Smout AJ. pH testing: the basics. J Clin Gastroenterol
6. Klausser AG, Schindlbeck NE, Müller-Lissner SA, et al. Esophageal 24 h Monitoring: is prior manometry
necessary for correct positioning of the electrode? Am J Gastroenterol
7. Murray JA, Clouse RE, Conklin JL, et al. Components of the standard oesophageal manometry
. Neurogastroenterol Motil
8. 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
9. Omari TI, Benninga MA, Haslam RR, et al. Lower esophageal sphincter position in premature infants cannot be correctly estimated with current formulas. J Pediatr
10. Emmerson AJB, Chant T, May J, et al. Assessment of three methods of pH probe positioning in preterm infants. J Pediatr Gastroenterol Nutr
11. Staiano A, Clouse R. Value of subject height in predicting lower esophageal sphincter location. Am J Dis Child
12. Working Group of the European Society of Pediatric Gastroenterology and NutritionStandardized protocol for the methodology of esophageal pH monitoring
and interpretation of the data for the diagnosis of gastroesophageal reflux. J Pediatr Gastroenterol Nutr
13. Dunne DP, Paterson WG. Acid-induced esophageal shortening in humans: a cause of hiatus hernia? Can J Gastroenterol
14. Ayazi S, Hagen JA, Zehetner J, et al. The value of high-resolution manometry
in the assessment of the resting characteristics of the lower esophageal sphincter. J Gastrointest Surg
15. Gilger MA, Boyle JT, Sondheimer JM, et al. A medical position statement of the North American Society for Pediatric Gastroenterology and Nutrition. Indications for pediatric esophageal manometry
. J Pediatr Gastroenterol Nutr