Abstract
This study assesses the performance of the simplified American Academy of Pediatrics table vs the American Academy of Pediatrics guideline for screening children with elevated blood pressure.
In 2017, the American Academy of Pediatrics (AAP) updated the clinical practice guideline for diagnosis and treatment of high blood pressure (BP) in children and adolescents (hereafter referred to as the AAP guideline).1 Similar to other pediatric BP guidelines,2,3 the AAP guideline is based on several hundreds of sex-, age-, and height-specific BP cutoffs, which is cumbersome to use in clinical practice. However, the AAP guideline also provides a simplified table based on the 90th BP percentiles at the 5th height percentile as a convenient tool to screen for elevated BP in children aged 1 to 12 years (hereafter referred to as the simplified AAP table).1 In a recent publication,4 we assessed simplified BP tables based on different height percentiles for screening elevated BP in children. In this study, we assessed the performance of the simplified AAP table vs the AAP guideline for screening children with elevated BP according to sex, race/ethnicity, BP status, and weight status.
Methods
We used data from 6816 children aged 8 to 12 years from 9 cycles of the National Health and Nutrition Examination Survey (NHANES5) conducted from 1999-2000 to 2015-2016 and from 3145 children aged 5 to 12 years from NHANES III conducted between 1988 and 1994. Adolescents aged 13 to 17 years were not included because the AAP guideline recommends only one BP cutoff for them (120/80 mm Hg). We used the average of the second and third BP readings for data analyses.
We compared the performance of the simplified AAP table to predict elevated BP as defined by the AAP guideline. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were estimated by receiver operating characteristic curve analysis. The NHANES data are freely open and available to the public; thus, institutional review board approval was not required.
Results
The prevalence of elevated BP was 12.1% in children aged 8 to 12 years in NHANES and 7.5% in children aged 5 to 12 years in NHANES III according to the AAP guideline. Comparing the simplified AAP table with the AAP guideline to predict elevated BP, sensitivity (99.9%) and NPV (100.0%) were optimal, but specificity was not optimal (84.4%), resulting in low PPV (46.9%) in NHANES (Table 1). Results were similar according to sex, race/ethnicity, BP status, and body mass index category. These indexes were similar using data from NHANES III (88.4% specificity and 45.1% PPV) (Table 2).
Table 1. Performance of the Simplified AAP Table to Predict Elevated BP as Defined by the AAP Guideline in Children in NHANES 1999-2016.
| Variable | % (95% CI) | |||
|---|---|---|---|---|
| Sensitivity | Specificity | PPV | NPV | |
| Total | 99.9 (99.8-100.0) | 84.4 (83.5-85.3) | 46.9 (45.7-48.1) | 100.0 (100.0-100.0) |
| Sex | ||||
| Boys | 99.8 (99.6-100.0) | 81.5 (80.2-82.8) | 44.6 (42.9-46.3) | 100.0 (100.0-100.0) |
| Girls | 100.0 (100.0-100.0) | 87.1 (86.0-88.2) | 49.8 (48.1-51.5) | 100.0 (100.0-100.0) |
| Race/ethnicity | ||||
| Hispanic | 100.0 (100.0-100.0) | 86.0 (84.6-87.4) | 50.2 (48.3-52.1) | 100.0 (100.0-100.0) |
| White | 99.6 (99.3-99.9) | 85.2 (83.6-86.8) | 47.2 (45.0-49.4) | 99.9 (99.8-100.0) |
| Black | 100.0 (100.0-100.0) | 81.2 (79.4-83.0) | 43.0 (40.7-45.3) | 100.0 (100.0-100.0) |
| Other | 100.0 (100.0-100.0) | 84.7 (81.8-87.6) | 46.0 (42.0-50.0) | 100.0 (100.0-100.0) |
| BP status | ||||
| Elevated SBP | 100.0 (100.0-100.0) | 85.7 (84.9-86.5) | 43.4 (42.2-44.6) | 100.0 (100.0-100.0) |
| Elevated DBP | 99.6 (99.5-99.7) | 97.2 (96.8-97.6) | 55.7 (54.5-56.9) | 100.0 (100.0-100.0) |
| BMI category | ||||
| Normal weight | 99.7 (99.5-99.9) | 89.6 (88.6-90.6) | 43.8 (42.2-45.4) | 100.0 (100.0-100.0) |
| Overweight or obesea | 100.0 (100.0-100.0) | 76.1 (74.6-77.6) | 48.6 (46.8-50.4) | 100.0 (100.0-100.0) |
Abbreviations: AAP, American Academy of Pediatrics; BMI, body mass index; BP, blood pressure; DBP, diastolic blood pressure; NHANES, National Health and Nutrition Examination Survey; NPV, negative predictive value; PPV, positive predictive value; SBP, systolic blood pressure.
Overweight or obese was defined using the 2000 US Centers for Disease Control and Prevention growth charts.6
Table 2. Performance of the Simplified AAP Table to Predict Elevated BP as Defined by the AAP Guideline in Children in NHANES III 1988-1994.
| Variable | % (95% CI) | |||
|---|---|---|---|---|
| Sensitivity | Specificity | PPV | NPV | |
| Total | 100.0 (100.0-100.0) | 88.4 (87.3-89.5) | 45.1 (43.4-46.8) | 100.0 (100.0-100.0) |
| Sex | ||||
| Boys | 100.0 (100.0-100.0) | 87.4 (85.8-89.0) | 44.8 (42.3-47.3) | 100.0 (100.0-100.0) |
| Girls | 100.0 (100.0-100.0) | 89.4 (87.9-90.9) | 45.6 (43.1-48.1) | 100.0 (100.0-100.0) |
| Race/ethnicity | ||||
| Hispanic | 100.0 (100.0-100.0) | 89.4 (87.6-91.2) | 49.5 (46.5-52.5) | 100.0 (100.0-100.0) |
| White | 100.0 (100.0-100.0) | 88.5 (86.4-90.6) | 42.3 (39.0-45.6) | 100.0 (100.0-100.0) |
| Black | 100.0 (100.0-100.0) | 86.8 (84.8-88.8) | 42.4 (39.4-45.4) | 100.0 (100.0-100.0) |
| Other | 100.0 (100.0-100.0) | 92.5 (88.2-96.8) | 52.4 (44.3-60.5) | 100.0 (100.0-100.0) |
| BP status | ||||
| Elevated SBP | 100.0 (100.0-100.0) | 90.7 (89.7-91.7) | 40.1 (38.4-41.8) | 100.0 (100.0-100.0) |
| Elevated DBP | 100.0 (100.0-100.0) | 96.3 (95.6-97.0) | 50.0 (48.3-51.7) | 100.0 (100.0-100.0) |
| BMI category | ||||
| Normal weight | 100.0 (100.0-100.0) | 91.8 (90.6-93.0) | 44.2 (42.1-46.3) | 100.0 (100.0-100.0) |
| Overweight or obesea | 100.0 (100.0-100.0) | 80.3 (77.8-82.8) | 46.5 (43.4-49.6) | 100.0 (100.0-100.0) |
Abbreviations: AAP, American Academy of Pediatrics; BMI, body mass index; BP, blood pressure; DBP, diastolic blood pressure; NHANES, National Health and Nutrition Examination Survey; NPV, negative predictive value; PPV, positive predictive value; SBP, systolic blood pressure.
Overweight or obese was defined using the 2000 US Centers for Disease Control and Prevention growth charts.6
Discussion
Because the simplified AAP table is based on BP cutoffs at a low height percentile (fifth) and because height is the major driver of BP in children, it is no surprise that the sensitivity and NPV of the simplified AAP table are optimal, resulting in virtually no false-negative cases. However, the trade-off for high sensitivity of the AAP guideline at a low height percentile was a low specificity (<89%) and a subsequently low PPV. With a PPV as low as 46.9%, less than half of all children with a positive test result actually had elevated BP. Tables of BP cutoffs based on low child height percentiles are bound to produce high rates of false-positive cases.7 Admittedly, BP readings of children who test positive with the simplified AAP table are expected to be further evaluated with the AAP guideline to confirm the presence of elevated BP.
It could be argued that a simplified table should be based on a higher height percentile for sex and age. This would result in higher specificity and PPV, with (most probably) only minimal influence on the sensitivity and NPV of the test. An alternate strategy could be to use a table of BP cutoffs based on height, which have been shown to have both high PPV and high NPV.7
Conclusions
Our analysis showed that the simplified AAP table had perfect sensitivity but limited specificity, resulting in large proportions of false-positive cases. Further research is needed to determine which screening tool would be easier to implement in routine clinical practice so more children are screened.
References
- 1.Flynn JT, Kaelber DC, Baker-Smith CM, et al. ; Subcommittee on Screening and Management of High Blood Pressure in Children . Clinical practice guideline for screening and management of high blood pressure in children and adolescents [published correction appears in Pediatrics. 2017;140(6)]. Pediatrics. 2017;140(3):e20171904. doi: 10.1542/peds.2017-1904 [DOI] [PubMed] [Google Scholar]
- 2.National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents . The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics. 2004;114(2)(suppl 4th report):555-576. doi: 10.1542/peds.114.2.S2.555 [DOI] [PubMed] [Google Scholar]
- 3.Xi B, Zong X, Kelishadi R, et al. ; International Child Blood Pressure References Establishment Consortium . Establishing international blood pressure references among nonoverweight children and adolescents aged 6 to 17 years. Circulation. 2016;133(4):398-408. doi: 10.1161/CIRCULATIONAHA.115.017936 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yang L, Hou Y, Zhao M, Bovet P, Xi B. Simplified blood pressure tables based on different height percentiles for screening elevated blood pressure in children [published online July 30, 2018]. J Hypertens. doi: 10.1097/HJH.0000000000001880 [DOI] [PubMed] [Google Scholar]
- 5.McQuillan GM, McLean JE, Chiappa M, Corporation H, Lukacs SL. National Health and Nutrition Examination Survey Biospecimen Program: NHANES III (1988-1994) and NHANES 1999-2014. Vital Health Stat 2. 2015;170(170):1-14. [PubMed] [Google Scholar]
- 6.Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11. 2002;246(246):1-190. [PubMed] [Google Scholar]
- 7.Ma C, Kelishadi R, Hong YM, et al. Performance of eleven simplified methods for the identification of elevated blood pressure in children and adolescents. Hypertension. 2016;68(3):614-620. doi: 10.1161/HYPERTENSIONAHA.116.07659 [DOI] [PMC free article] [PubMed] [Google Scholar]
