Abstract
Background
Children undergoing cardiac surgery (CS) are at risk of high blood pressure (BP), a risk factor for cardiovascular and kidney disease. 24-hour ambulatory BP monitoring (ABPM) is a reference standard hypertension (HTN) test. Little data exist on ABPM abnormalities in children several years post-CS.
Objectives
Nine years post-CS: a) determine ABPM feasibility; b) describe ABPM abnormality prevalence (percent load; masked HTN [MH]; non-dipping, mean systolic/diastolic BP [SBP/DBP] >95th percentile), pre-HTN (ABPM), white coat HTN [WCH]; c) the relationship between ABPM and casual BP; d) compare BP in patients with and without acute kidney injury (AKI).
Methods
Prospective, longitudinal pilot study of children (0–18 years) who underwent CS from 2007–2009 at Montreal Children’s Hospital (TRIBE-AKI sub-cohort). Measurements: peri-operative AKI; BP classified by three single-visit measures (normal; elevated BP [eBPSingleVisit]; HTNSingleVisit); ABPM. Univariable analyses were used to compare characteristics between groups.
Results
ABPM feasibility: Twenty-four participants performed ABPM; 23 were included in analysis (consent rate: 76%; ABPM success rate: 86%). Six participants (26%) had eBPSingleVisit or higher. ABPM abnormalities: 3 (13%) pre-HTN; 0 MH; 9 (39%) non-dipping; 11 (48%) ≥1 ABPM abnormality. Three (13%) had WCH. In 8 AKI patients, none had pre-HTN, MH, or WCH. In 15 non-AKI patients, 3 (20%), 0, and 3 (20%) had pre-HTN, MH, and WCH respectively.
Conclusion
9 years post-CS, ABPM was feasible, with frequent ABPM abnormalities but low ABPM-defined HTN prevalence. AKI was not associated with worse BP outcomes. Research on BP risk factors and guidelines after CS is needed to reduce cardiovascular risk.
INTRODUCTION
We previously showed in the Translational Research Investigating Biomarker End Points in Acute Kidney Injury (AKI) 5 year follow-up study (TRIBE-AKI 2), that 17% children with congenital heart defects who underwent cardiac surgery had hypertension (HTN), ascertained by three blood pressure (BP) measures during a single study visit (hereafter referred to as HTNSingleVisit)1. Given this high prevalence of HTNSingleVisit (over 8 times that of the general pediatric population, assessed using similar measures)2, confirming these findings using 24-hour ambulatory blood pressure monitoring (ABPM) may have significant implications on estimating cardiovascular disease risk after pediatric cardiac surgery. ABPM would also elucidate the prevalence of specific ABPM abnormalities, including ambulatory HTN, masked HTN (MH), white-coat HTN (WCH) and non-dipping, several of which may be associated with measures of cardiovascular risk in different pediatric populations3. Understanding the disparity between ABPM and casual BP in children who have undergone cardiac surgery, may impact recommendations for follow-up care and advise on how best to perform long-term BP research in this patient population. Most studies evaluating BP outcomes using ABPM in this population have been limited to patients in specific cardiac surgery contexts (e.g., heart transplant) and with small sample sizes4,5. There is thus a considerable knowledge gap on long-term BP outcomes in children undergoing cardiac surgery. Given the heterogeneity of both cardiac diagnoses and illness severity levels of children undergoing cardiac surgery, research on long-term (>5 years) ABPM abnormalities will require large sample sizes and incur considerable cost. Understanding the feasibility (e.g., ABPM success rate; consent to perform the test when it is not part of routine care) and challenges of performing ABPM in this patient population will allow for anticipation of problems and development of solutions to overcome these challenges6.
The potential pathophysiology of HTN development in children undergoing cardiac surgery is likely multifactorial, but may include chronic hypoxia with microvascular changes, chronic kidney damage (from polycythemia, hypoxia or poor kidney blood flow, with neurohormonal aberrations) or obesity development7,8. We previously evaluated the association of post-operative AKI with 5-year post-cardiac surgery HTN, hypothesizing that AKI leads to kidney and possibly microvascular damage, with resulting HTN; however, we did not find an association1. Because we only evaluated BP during a single visit in that study, it is possible that using a better reference standard, such as ABPM, an association with AKI might be detected if present.
The primary objectives of this pilot study were to describe the feasibility of performing ABPM for research at approximately 9 years after pediatric cardiac surgery and evaluate the relation between ABPM abnormalities and casual BP. We hypothesized that it would be challenging to perform ABPM in this patient population who are high-risk for BP abnormalities, but also at a considerable number of years after their initial surgery. We also hypothesized that there would be discordance between casual BP and ABPM measures. A secondary objective was to evaluate ABPM abnormalities in patients with vs. without peri-operative AKI.
METHODS
Study cohort
This was a prospective longitudinal cohort pilot study conducted in children who underwent cardiac surgery at the Montreal Children’s Hospital between the ages of 1 month to 18 years old. The children were originally enrolled pre-operatively into the TRIBE-AKI 1 study, a three-centre study occurring between July 2007 and December 2009 to study biomarkers of peri- operative AKI9. They were subsequently enrolled into a 5 year follow-up study (from the index cardiac surgery) of kidney and BP outcomes (TRIBE-AKI 2; details on recruitment published elsewhere)1. For the current pilot study, we only included Montreal site participants and excluded participants who refused or were unable to provide consent or assent (if > 7 years old) or who lived approximately ≥3.5 hours drive (~120 miles) away and could not come to the study center to perform study activities. For the analysis, we excluded participants who did not have complete wake and sleep ABPM readings or who were greater than 18 years old. This study was approved by the Research Ethics Board of the Montreal Children’s Hospital and parental consent and/or child assent were obtained prior to initiating any study activities.
Recruitment and study visits
After the 5 year follow-up TRIBE-AKI 2 study, participants were contacted by phone each year, to maintain contact. Near the 9 year post-cardiac surgery follow-up time point, Montreal participants were invited to participate in the current study.
In-person study visits were conducted at approximately 9 years post-cardiac surgery by a research nurse from the Montreal Children’s Hospital, either at the hospital center or in the participants’ homes (as previously described)1. Blood and urine samples were collected and then shipped and stored at Yale University (the original TRIBE-AKI primary study site). Three height and weight measurements were performed and used to calculate body mass index (BMI) and percentiles of height, weight and BMI10. Clinical data collection included medical history updates (including diagnoses, anti-hypertensive medications, renal and non-renal history).
Casual BP measurement and classification
During the single study visit, three casual BP measures were performed using the Omron HEM-711ac BP monitor (Omron Healthcare, Inc), in a calm setting (at the participant’s home or a quiet room in the study centre), seated, in the right arm at least 3 minutes apart, and using a size-appropriate cuff. If a BP reading showed an error message, BP was measured by auscultation. The lowest two systolic BP (SBP) measures and the corresponding diastolic BP (DBP) measures were averaged and used to calculate age, sex, and height-adjusted BP percentiles based on the 2017 American Academy of Pediatrics Guidelines (software at https://apps.cpeg-gcep.net/BPz_cpeg/) or to define BP abnormalities based on threshold values in older children, as per the guidelines11. Guidelines state that repeated measures from different days are required to diagnose BP abnormalities by casual BP11; however this was not feasible in the context of this study. To acknowledge this, the nomenclature used to classify participants’ casual BP measured during this single visit was: normal BPSingleVisit, elevated BP (eBPSingleVisit), Stage 1 HTNSingleVisit or Stage 2 HTNSingleVisit, using calculated BP percentiles and/or BP level thresholds, as stated in the 2017 American Academy of Pediatrics Guidelines11. The presence of high casual BPSingleVisit hereafter refers to presence of eBPSingleVisit or worse.
ABPM measurement and classifications
The ABPM was performed using the Spacelabs 90217A device (Issaquah, WA), over 24 hours, with a goal of BP readings every 30 minutes. A minimum of 70% valid readings was considered sufficient. Wake and sleep periods were based on sleep and wake times recorded in the participant’s journal. Normative ABPM BP data presented in the American Heart Association Guidelines and associated references were used to define ABPM abnormal BP threshold values and percentiles and to define ABPM-specific BP abnormalities3,12. BP percent load was defined as the percentage of readings above the 95th percentile for a given period (i.e., wake or sleep). Elevated BP percent load was defined as ≥ 25% of readings above the 95th percentile for wake and/or sleep periods. SBP or DBP dipping was the percent difference of mean sleep to wake SBP and DBP (i.e., [mean awake BP mmHg- mean sleep BP mmHg]/mean awake BP mmHg × 100). Non-dipping was SBP or DBP dipping < 10%. Presence of mean ABPM-measured SBP or DBP > 95th percentile was determined. ABPM-defined BP index (not a measure included in the ABPM guidelines) was defined as the mean BP (for wake or sleep, SBP and DBP) divided by 95th percentile cut-off (BP index value >1 was classified as abnormal, as previously described)13.
As per the American Heart Association Guidelines, the following ABPM classifications were ascertained: normal BPSingleVisit was classified as a casual BPSingleVisit < 90th percentile and a mean BP by ABPM < 95th percentile without elevated BP percent load; WCH was a casual BPSingleVisit ≥ 95th percentile with a mean BP by ABPM < 95th percentile and without elevated BP percent load; MH was a casual BPSingleVisit < 95th percentile and a mean BP by ABPM > 95th percentile with an elevated BP percent load; Pre-HTN (by ABPM) was a casual BPSingleVisit ≥ 90th percentile or >120/80 mmHg and a mean BP by ABPM < 95th percentile with an elevated BP percent load; ambulatory HTN was a casual BPSingleVisit > 95th percentile and a mean BP by ABPM > 95th percentile and an elevated BP load between 25–50% (if BP load >50%, termed severe ambulatory HTN)3.
Index hospitalization data and 9 year follow-up laboratory data
Information from the index cardiac surgery (including pre-operative characteristics, operative details, and post-operative complications) were available from our previously performed studies1,9. The risk adjustment for congenital heart surgery-1 (RACHS-1) score (widely accepted to evaluate differences in outcomes for congenital heart disease) was used to categorize surgical complexity (higher score associated with more complex surgeries)14. Post-operative AKI was defined as development of ≥stage 1 AKI, defined as a ≥ 50% or a ≥ 0.3 mg/dL (or 26.5 μmol/L) post-operative increase in serum creatinine from pre-operative baseline (available for all participants)15. Pre-operative estimated glomerular filtration rate (eGFR) was calculated using the Chronic kidney Disease in Children study (CKiD) serum-creatinine based bedside equation16.
From the day of the 9 year follow-up study visit, serum creatinine (isotope dilution mass spectrometry traceable assay) was used to calculate eGFR. Urine albumin to creatinine ratio (from a spot urine specimen) was used to ascertain microalbuminuria (urine albumin to creatinine ratio > 30 mg/g)17. Chronic kidney disease at the 9 year follow-up was defined as eGFR < 90 mL/min/1.73 m2 or microalbuminuria17.
Statistical analysis
Continuous variables were expressed as median (interquartile range) [IQR] and categorical variables were expressed as frequencies and percentages. The Mann Whitney U-test was used to compare continuous variables between groups, and Chi-square or Fisher’s exact test were used to compare categorical variables, as appropriate. Feasibility was assessed by consent rate and by proportion of patients with sufficient wake and/or sleep ABPM measures. The relationship between ABPM and casual BPSingleVisit was assessed by comparing ABPM abnormalities in groups of high casual BPSingleVisit (eBPSingleVisit; stage 1 HTNSingleVisit and stage 2 HTNSingleVisit) vs. normal casual BPSingleVisit.
AKI vs. non-AKI groups were compared by comparing proportions with high casual BPSingleVisit and by comparing proportions of ABPM abnormalities by AKI group. Analyses were conducted using STATA® version 12 (College Station, TX, USA).
RESULTS
Study cohort and ABPM feasibility
Forty-five of the 131 participants enrolled into the 5 year follow-up TRIBE-AKI 2 study were from the Montreal site. Three participants were lost to follow-up, three lived too far from the study centre and one had died; 28 of the 38 available participants (74%) consented to the study. Of these 28 participants, 24 (86%) had successful wake and sleep ABPM readings; one was excluded from analysis due to age>18 years old, leaving 23 participants in the final analysis population. Characteristics between patients included in the analysis (n=23) vs. those excluded (n=5) were not significantly different (Online Resource 1).
Participant characteristics and 9 year follow-up casual BP
A summary of baseline, surgical, and 9 year follow-up characteristics for the 23 participants is shown in Table 1. Post-operative AKI occurred in 8 (35%) participants. Six (26%) participants had a surgery RACHS-1 category ≥3. Three (13%) participants underwent an aortic coarctation repair (one had high casual BPSingleVisit and received anti-hypertensive medication; one received anti-hypertensive medication with normal casual BPSingleVisit; one had normal casual BPSingleVisit and was not taking anti-hypertensive medication).
Table 1.
Characteristics of study cohort
| Characteristics | Overall (n= 23) |
|---|---|
| Index Hospitalization | |
| Age (years) at Index Cardiac Curgery | 2.4 (0.6, 7.9) |
| Male | 16 (70%) |
| Non-White | 4 (17%) |
| eGFR ml/min/1.73m2 (Pre-operative) | 73.3 (52.6, 85.0) |
| Past cardiac surgeries | 5 (22%) |
| CPB time (min) | 79.0 (52.0, 120.0) |
| Type of surgery | |
| Septal Defect Repair | 3 (13%) |
| Inflow/ outflow tract or valve procedure | 11 (48%) |
| Combined Procedure | 4 (17%) |
| Other | 5 (22%) |
| RACHS-1 Category 1 | 2 (9%) |
| RACHS-1 Category 2 | 15 (65%) |
| RACHS-1 Category 3 | 5 (22%) |
| RACHS-1 Category 4 | 1 (4%) |
| AKI | 8 (35%) |
| Stage 1 | 7 (30%) |
| Stage 2 | 1 (5%) |
| Length of Hospital stay (days) | 6.0 (5.0, 15.0) |
| Length of ICU stay (days) | 2.0 (1.0, 4.0) |
| 9 year Follow-up Visit | |
| Age (years) at 9 year follow up | 10.8 (9.1, 13.8) |
| Years Post Cardiac Surgery | 8.6 (8.0, 9.0) |
| Height (Percentile) | 49.6 (12.3, 85.1) |
| Weight (Percentile) | 57.1 (34.8, 86.9) |
| BMI (Percentile) | 70.2 (16.5, 92.9) |
| Overweight | 6 (26%) |
| Obese | 2 (9%) |
| Severely Obese | 2 (9%) |
| †eGFR (ml/min/1.73m2) at 9 year follow up | 107.7 (88.4, 129.3) |
| †CKD | 5/17 (29%) |
| Taking Anti-Hypertensive Medications | 5 (22%) |
| Casual BP classifications | |
| Normal BPSingleVisit | 17 (74%) |
| Elevated BP (eBP)SingleVisit | 2 (9%) |
| Stage 1 HTNSingleVisit | 3 (13%) |
| Stage 2 HTNSingleVisit | 1 (4%) |
Values are presented as frequency (percentage) or as the median, interquartile range (IQR).
Based on n=17 patients who had blood and urine collected.
Abbreviations: CPB, cardiopulmonary bypass; AKI, Acute kidney injury; eGFR, estimated glomerular filtration rate; RACHS-1, risk-adjustment for congenital heart surgery; ICU, intensive care unit; BMI, Body mass index; CKD, Chronic kidney disease.
At 9 year follow-up, 6 (26%) participants had high casual BPSingleVisit: 2 participants had casual eBPSingleVisit, 3 had stage 1 HTNSingleVisit, and 1 had stage 2 HTNSingleVisit. Five (22%) participants were taking anti-hypertensive medication: 2 (40%) of these had high casual BPSingleVisit (casual eBPSingleVisit or higher) at the study visit .
ABPM abnormalities at 9 year follow-up and relation with casual BPSingleVisit
In 23 participants, the prevalence of WCH and MH was 13% (n=3) and 0%, respectively; 3 (13%) had pre-HTN (ABPM) and no participants had ambulatory HTN. In the 3 participants with WCH, 1 was taking anti-hypertensive medication. For the 3 participants with pre-HTN (ABPM), all three fulfilled pre-HTN criteria during the wake period; only one also fulfilled criteria during the sleep period.
During the wake period of ABPM, mean SBP and DBP percentiles and BP percent load were significantly higher in participants who did vs. did not have high casual BPSingleVisit (Table 2). For the sleep ABPM data, the relations between ABPM BP percentiles and percent load with high casual BPSingleVisit were in a similar direction, but only sleep SBP percent load was significantly higher in the high casual BPSingleVisit group (Table 2). Prevalence of non-dipping in the whole study cohort was 39%; however, the prevalence was higher in the normal vs. high casual BPSingleVisit group (47% vs. 17%, respectively, Table 2, not statistically significant). When considering all ABPM abnormalities, 11 (48%) had at least one wake or sleep ABPM abnormality (consisting of either pre-HTN [ABPM], elevated SBP or DBP percent load or non-dipping; at least one abnormality present in 8/17 (47%) participants with normal vs. 3/6 (50%) participants with high casual BPSingleVisit). Online Resource 2 shows that the 5 participants taking anti-hypertensive medication (vs. not taking anti-hypertensive medication) generally had higher wake period BP, but had significantly higher sleep period percent SBP dipping (17% vs. 10% SBP dipping, p=0.002, Online Resource 2).
Table 2.
ABPM results (9 years post cardiac surgery) comparing patients with normal casual BPSingleVisit to high casual BPSingleVisit.
| ABPM Variables/ Abnormalities | Total | Normal Casual BPSingleVisit | High Casual BPSingleVisita |
|---|---|---|---|
| Wake ABPM Period | (n=23) | (n=17) | (n=6) |
| Mean SBP %tile | 14.3 (2.5, 45.0) | 12.7 (2.0, 22.2) | *60.8 (38.8, 76.8) |
| Mean DBP %tile | 11.0 (6.2, 22.0) | 6.9 (5.7, 12.3) | *17.5 (12.8, 37.9) |
| SBP % load | 0.0 (0.0, 9.5) | 0.0 (0.0, 3.7) | *13.0 (4.2, 25.9) |
| DBP % load | 3.7 (0.0, 4.8) | 3.7 (0.0, 4.5) | 3.9 (3.2, 7.4) |
| Elevated SBP and/ or DBP % load | 3 (13%) | 0 (0%) | *3 (50%) |
| Mean SBP or DBP > 95%tile | 0 (0%) | 0 (0%) | 0 (0%) |
| Mean SBP Index | 0.8 (0.8, 0.9) | 0.8 (0.8, 0.9) | 0.9 (0.8, 0.9) |
| Mean DBP Index | 0.8 (0.8, 0.8) | 0.8 (0.8, 0.8) | 0.8 (0.8, 0.9) |
| Sleep ABPM Period | (n=23) | (n=17) | (n=6) |
| Mean SBP %tile | 29.2 (13.7, 47.2) | 22.1 (13.7, 32.4) | 41.7 (29.6, 73.8) |
| Mean DBP %tile | 27.9 (13.6, 39.5) | 17.0 (13.6, 29.5) | 37.9 (25.0, 40.0) |
| SBP % load | 0.0 (0.0, 5.0) | 0.0 (0.0, 0.0) | *7.5 (0.0, 21.4) |
| DBP % load | 0.0 (0.0, 6.3) | 0.0 (0.0, 6.3) | 0.0 (0.0, 5.9) |
| Elevated SBP and/ or DBP % load | 2 (9%) | 1 (6%) | 1 (17%) |
| Mean SBP or DBP >95%ile | 1 (4%) | 0 (0%) | 1 (17%) |
| Mean SBP Index | 0.8 (0.8, 0.9) | 0.8 (0.8, 0.9) | 0.9 (0.9, 0.9) |
| Mean DBP Index | 0.8 (0.8, 0.8) | 0.8 (0.8, 0.8) | 0.8 (0.7, 0.8) |
| % dipping SBP | 10.7 (7.7, 15.0) | 10.4 (7.2, 14.4) | *16.0 (11.7, 19.1) |
| % dipping DBP | 19.0 (16.1, 22.1) | 19.4 (17.3, 21.8) | 23.4 (18.1, 30.6) |
| Non-dipping (SBP and/ or DBP) | 9 (39%) | 8 (47%) | 1 (17%) |
Values are presented as frequency (percentage) or as the median (IQR).
p<0.05.
Casual BP groups are based on Casual BP measured during a single visit.
Abbreviations: SBP: Systolic BP; DBP: Diastolic BP.
High casual BP defined by presence of eBPSingleVisit, stage 1 HTNSingleVisit or stage 2 HTNSingleVisit during casual BP measurement during a single visit.
AKI association with 9 year follow-up ABPM
Eight (35%) participants had a history of post-operative AKI at the index cardiac surgery hospitalization. Online Resource 3 shows that the index hospitalization and 9 year follow-up visit characteristics (including anthropometric measures and eGFR) were not significantly different between AKI groups (cardiac surgery procedures by RACHS-1 surgical category shown in Online Resource 4). All 5 participants fulfilling our definition for CKD (Online Resource 3) and all 6 participants with high casual BPSingleVisit were in the non-AKI group. None of the 8 participants with AKI had pre-HTN, MH, or WCH; 3/15 (20%) non-AKI participants had pre-HTN and 3 (20%) had WCH. Table 3 shows that in general, AKI participants had lower ABPM-measured BP percentiles for wake and sleep periods compared to non-AKI patients, however differences were not statistically significant (Table 3).
Table 3:
ABPM results (9 years post cardiac surgery) comparing patients with vs. without post-operative AKI.
| ABPM Variables/ Abnormalities | AKI patients | Non-AKI patients |
|---|---|---|
| Wake ABPM Period | (n=8) | (n=15) |
| Mean SBP %tile | 6.6 (3.4, 21.4) | 22.2 (2.5, 55.5) |
| Mean DBP %tile | 7.3 (6.2, 12.2) | 12.8 (5.7, 31.2) |
| SBP % load | 0.0 (0.0, 1.9) | 3.7 (0.0, 12.0) |
| DBP % load | 1.9 (0.0, 4.4) | 3.7 (0.0, 7.4) |
| High SBP and/ or DBP % load | 0 (0%) | 3 (20%) |
| Mean SBP or DBP > 95%tile | 0 (0%) | 0 (0%) |
| Mean SBP Index | 0.8 (0.8, 0.8) | 0.9 (0.8, 0.9) |
| Mean DBP Index | 0.8 (0.7, 0.8) | 0.8 (0.8, 0.8) |
| Sleep ABPM Period | (n=8) | (n=15) |
| Mean SBP %tile | 22.4 (9.1, 29.8) | 32.4 (16.3, 49.9) |
| Mean DBP %tile | 21.6 (12.7, 29.7) | 28.9 (14.1, 40.0) |
| SBP % load | 0.0 (0.0, 2.5) | 0.0 (0.0, 6.3) |
| DBP % load | 0.0 (0.0, 6.9) | 0.0 (0.0, 6.3) |
| High SBP and/ or DBP % load | 0 (0%) | 1 (7%) |
| Mean SBP or DBP >95%ile | 1(13%) | 0 (0%) |
| Mean SBP Index | 0.8 (0.8, 0.8) | 0.9 (0.8, 0.9) |
| Mean DBP Index | 0.7 (0.8, 0.8) | 0.8 (0.8, 0.8) |
| % dipping SBP | 12.3 (7.7, 14.6) | 11.7 (7.7, 16.3) |
| % dipping DBP | 19.6 (16.2, 21.4) | 21.1 (17.3, 25.6) |
| Non-dipping (SBP and/ or DBP) | 3 (38%) | 6 (40%) |
Values are presented as frequency (percentage), or the median (IQR), as appropriate.
Abbreviations: AKI, Acute kidney injury; SBP, Systolic BP; DBP, Diastolic BP.
DISCUSSION
We showed that consent and completion rates for ABPM in a research setting were 74% and 86%, respectively, suggesting that larger scale long-term pediatric cardiac surgery ABPM outcome studies are feasible. Acknowledging our small sample size, the 9 year prevalence of ABPM-defined HTN was low, but individual ABPM abnormalities were common, suggesting a need to understand cardiovascular risk impact of these abnormalities in children undergoing cardiac surgery.
Evaluating ABPM feasibility in this population is important. Little data on ABPM in children undergoing cardiac surgery have been published, children with congenital cardiac defects are at risk for long term cardiac events8 and recruiting patients with substantial past medical and surgical history into studies requiring non-routinely collected measures may be quite challenging6. The consent rate we achieved may be higher than that from a general pediatric cardiac surgery population, since our cohort previously participated in studies and might be amenable to research.
We previously showed that at a single study visit performed 5 years after cardiac surgery in this cohort, prevalence of BP in the HTN category was 17%1, similar to what we found about 4 years later in this study. This prevalence is high and sustained, in comparison to the prevalence of childhood HTN of <1% (<3% for elevated BP), from a national Canadian cohort (also measured during a single visit)2. Considering that not all participants with abnormal casual BP were being treated with anti-hypertensive medication, this is concerning, given the long-term cardiovascular risk of this population7,8. However, no participants had MH or ambulatory HTN and 3 participants fulfilled criteria for pre-HTN by ABPM; 3 participants had WCH (one of whom was taking medication). These findings, which should be interpreted with caution given our sample size, have several implications. Future studies on ABPM-defined HTN in these patients will require large sample sizes to detect overt HTN by ABPM. Secondly, ABPM in these patients may be useful to rule out casual BP-measured HTN; this is relevant for avoiding unneeded anti-hypertensive medication that may cause complications.
ABPM provides more information than simply ascertaining HTN, consisting of a collection of BP-related data. In fact, 48% of our participants had ≥1 ABPM abnormality, mostly consisting of non-dipping. Non-dipping is associated with several secondary causes of HTN3,11, but the extent to which it is associated with cardiovascular risk factors or outcomes in children is controversial18–20. Non-dipping may, however, be associated with future risk for HTN or worsening BP21. This is intriguing, because in our cohort, participants with normal casual BPSingleVisit had worse sleep period dipping patterns. Future ABPM research in the pediatric cardiac surgery population should focus on determining if currently non-actionable ABPM abnormalities (e.g., pre-HTN, non-dipping, elevated load) are associated with surrogate cardiovascular outcomes (e.g., left ventricular hypertrophy, carotid intima media thickness) and whether treatment of these entities may be of benefit. Our study suggests event rates for these ABPM abnormalities which would be amenable to planning a study with a feasible sample size.
Questions which remain unanswered in the literature pertain to why children undergoing cardiac surgery might be at risk for long-term HTN and what drives the pathophysiology. Some cardiac defects are known to be associated with HTN, such as aortic coarctation (we had 3 such patients in our cohort). Other mechanisms underlying post-surgical HTN development include chronic kidney, vascular and neurohormonal effects from acute (peri-operative) and/or chronic hypoxia and cardiac structure and/or receptors dysfunction7,8. Post-cardiac surgery AKI has been proposed to be a risk factor for long-term kidney disease in children. However, this still remains controversial1,22. In this study, participants who did not have AKI had worse ABPM characteristics; this is consistent with our casual BP findings in our previous TRIBE-AKI 2, 5 year follow-up study1. A possible partial explanation could be that AKI patients also have worse chronic cardiac function with low output, manifesting as lower BP, but we were not able to evaluate this. There may be other confounders of the relation between AKI and later HTN development (including medications, adiposity, cyanosis, age groups) which we could not account for in this pilot study but should be elucidated in future research.
The strengths of this study include the prospective design, relatively prolonged follow-up time and extensive measures to achieve participant retention. There were limitations. Given the small sample size and pilot/feasibility nature of the study, findings must be interpreted with caution and larger studies are needed. Due to feasibility issues, we measured casual BP, using an automated device, over one day. Casual BP abnormalities may thus have been overestimated. We also did not perform echocardiogram to evaluate BP associated with end organ damage or cardiac function. Future studies should consider including echocardiographic outcome data. Given the potential lack of association of AKI with later HTN development, future studies should consider evaluating other factors (e.g., perinatal data like birth weight, gestational age or NICU admission; adiposity; HTN family history/genetics) which our study lacked.
This study highlights that there may be an important role for systematically monitoring children who have had cardiac surgery with ABPM, particularly if casual BP is abnormal. More data on post-pediatric cardiac surgery ABPM abnormalities and associations with clinical outcomes and other cardiometabolic risk factors, are needed. Our study suggests that such research is feasible, but may need to focus on sub-clinical ABPM abnormalities.
Acknowledgements
A special thanks also goes to the Research Nurse Julie Ann Doucet for help in performing study visits.
Funding/Support:
This study was supported by the National Institutes of Health (NIH) (grant R01HL085757 to Dr Parikh) to fund the TRIBE-AKI Consortium to study novel biomarkers of acute kidney injury in cardiac surgery. Dr Greenberg is funded by the NIH career development grant K08DK110536 and a Charles H. Hood Foundation grant. Dr Devarajan is supported by the NIH (grant P50DK096418). Dr Parikh is also a member of the NIH-sponsored Assess, Serial Evaluation, and Subsequent Sequelae in Acute Kidney Injury Consortium (U01DK082185). MZ was supported by a salary award from the Fonds de Recherche du Quebec - Sante (FRQ-S) during the majority of the conduct of this study.
Role of the Funder/Sponsor:
The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
REFERENCES
- 1.Greenberg JH, Zappitelli M, Devarajan P, Thiessen-Philbrook HR, Krawczeski C, Li S, Garg AX, Coca S, Parikh CR, Consortium T-A: Kidney Outcomes 5 Years After Pediatric Cardiac Surgery: The TRIBE-AKI Study. JAMA Pediatr 170:1071–1078, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Paradis G, Tremblay MS, Janssen I, Chiolero A, Bushnik T: Blood pressure in Canadian children and adolescents. Health Rep 21:15–22, 2010 [PubMed] [Google Scholar]
- 3.Flynn JT, Daniels SR, Hayman LL, Maahs DM, McCrindle BW, Mitsnefes M, Zachariah JP, Urbina EM, American Heart Association Atherosclerosis H, Obesity in Youth Committee of the Council on Cardiovascular Disease in the Y: Update: ambulatory blood pressure monitoring in children and adolescents: a scientific statement from the American Heart Association. Hypertension 63:1116–1135, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tainio J, Qvist E, Miettinen J, Holtta T, Pakarinen M, Jahnukainen T, Jalanko H: Blood pressure profiles 5 to 10 years after transplant in pediatric solid organ recipients. J Clin Hypertens (Greenwich) 17:154–161, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Luitingh TL, Lee MGY, Jones B, Kowalski R, Weskamp Aguero S, Koleff J, Zannino D, Cheung MMH, d’Udekem Y: A Cross-Sectional Study of the Prevalence of Exercise-Induced Hypertension in Childhood Following Repair of Coarctation of the Aorta. Heart Lung Circ 28:792–799, 2019 [DOI] [PubMed] [Google Scholar]
- 6.Rajadhyaksha V: Conducting feasibilities in clinical trials: an investment to ensure a good study. Perspect Clin Res 1:106–109, 2010 [PMC free article] [PubMed] [Google Scholar]
- 7.Morgan C, Al-Aklabi M, Garcia Guerra G: Chronic kidney disease in congenital heart disease patients: a narrative review of evidence. Can J Kidney Health Dis 2:27, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Roche SL, Silversides CK: Hypertension, obesity, and coronary artery disease in the survivors of congenital heart disease. Can J Cardiol 29:841–848, 2013 [DOI] [PubMed] [Google Scholar]
- 9.Li S, Krawczeski CD, Zappitelli M, Devarajan P, Thiessen-Philbrook H, Coca SG, Kim RW, Parikh CR: Incidence, risk factors, and outcomes of acute kidney injury after pediatric cardiac surgery: a prospective multicenter study. Crit Care Med 39:1493–1499, 2011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Prevention CfDCa 2017, June 16 Clinical Growth Charts.
- 11.Flynn JT, Kaelber DC, Baker-Smith CM, Blowey D, Carroll AE, Daniels SR, de Ferranti SD, Dionne JM, Falkner B, Flinn SK, Gidding SS, Goodwin C, Leu MG, Powers ME, Rea C, Samuels J, Simasek M, Thaker VV, Urbina EM: Clinical Practice Guideline for Screening and Management of High Blood Pressure in Children and Adolescents. Pediatrics 2017 [DOI] [PubMed] [Google Scholar]
- 12.Wuhl E, Witte K, Soergel M, Mehls O, Schaefer F, German Working Group on Pediatric H: Distribution of 24-h ambulatory blood pressure in children: normalized reference values and role of body dimensions. J Hypertens 20:1995–2007, 2002 [DOI] [PubMed] [Google Scholar]
- 13.Barletta GM, Pierce C, Mitsnefes M, Samuels J, Warady BA, Furth S, Flynn J: Is Blood Pressure Improving in Children With Chronic Kidney Disease? A Period Analysis. Hypertension 71:444–450, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jenkins KJ, Gauvreau K, Newburger JW, Spray TL, Moller JH, Iezzoni LI: Consensus-based method for risk adjustment for surgery for congenital heart disease. J Thorac Cardiovasc Surg 123:110–118, 2002 [DOI] [PubMed] [Google Scholar]
- 15.Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group. KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney inter, Suppl. 1–138, 2012 [Google Scholar]
- 16.Schwartz GJ, Munoz A, Schneider MF, Mak RH, Kaskel F, Warady BA, Furth SL: New equations to estimate GFR in children with CKD. J Am Soc Nephrol 20:629–637, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kidney Disease: Improving Global Outcomes CKDMBDWG: KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD). Kidney Int Suppl S1–130, 2009. [DOI] [PubMed] [Google Scholar]
- 18.Atabek ME, Akyurek N, Eklioglu BS, Alp H: Impaired systolic blood dipping and nocturnal hypertension: an independent predictor of carotid intima-media thickness in type 1 diabetic patients. J Diabetes Complications 28:51–55, 2014 [DOI] [PubMed] [Google Scholar]
- 19.Seeman T, Hradsky O, Gilik J: Nocturnal blood pressure non-dipping is not associated with increased left ventricular mass index in hypertensive children without end-stage renal failure. Eur J Pediatr 175:1091–1097, 2016 [DOI] [PubMed] [Google Scholar]
- 20.Westerstahl M, Hedvall Kallerman P, Hagman E, Ek AE, Rossner SM, Marcus C: Nocturnal blood pressure non-dipping is prevalent in severely obese, prepubertal and early pubertal children. Acta Paediatr 103:225–230, 2014 [DOI] [PubMed] [Google Scholar]
- 21.Deja G, Borowiec M, Fendler W, Pietrzak I, Szadkowska A, Machnica L, Polanska J, Mlynarski W, Jarosz-Chobot P: Non-dipping and arterial hypertension depend on clinical factors rather than on genetic variability of ACE and RGS2 genes in patients with type 1 diabetes. Acta Diabetol 51:633–640, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Madsen NL, Goldstein SL, Froslev T, Christiansen CF, Olsen M: Cardiac surgery in patients with congenital heart disease is associated with acute kidney injury and the risk of chronic kidney disease. Kidney Int 92:751–756, 2017 [DOI] [PubMed] [Google Scholar]