This study was designed to explore the value of flow-mediated vasodilation (FMD) as a predictor of therapeutic response to midodrine hydrochloride (MD) in children with postural orthostatic tachycardia syndrome (POTS). One hundred and eight children diagnosed with POTS and 20 healthy control children were enrolled. All children with POTS received MD and were followed up for 3 months. FMD of brachial artery for each participant was measured by vascular ultrasound. Symptom scores, FMD values, and head-up test (HUT)/head-up tilt test (HUTT) outcomes were investigated before and after treatment. A receiver operating characteristic curve was used to explore the value of FMD as a predictor. Baseline FMD (%) and increased heart rate (beats per minute) during HUT/HUTT were significantly greater in children with POTS compared with control children (FMD: 11 ± 3% vs 6 ± 2%, p <0.001; increased heart rate: 38 ± 9 vs 7 ± 7 beats/min, p <0.001, respectively). Before treatment, MD responders had greater FMD values than MD nonresponders (p <0.05). Symptom scores, excessive increases in heart rate during HUT, and increased FMD values were all reduced significantly after treatment (all p <0.05). The receiver operating characteristic curve for the predictive value of FMD showed the area under the curve to be 0.790 (95% confidence interval: 0.679 to 0.902; p <0.001) at 1-month and 0.803 (95% confidence interval: 0.669 to 0.936; p <0.01) at 3-month therapy. FMD of 9.85% had a high sensitivity (1-month therapy: 71.6%; 3-month therapy: 74.4%) and specificity (1-month therapy 77.8%; 3-month therapy: 80%). In conclusion, FMD is a predictor of the efficacy of MD for treating children with POTS.
The exact mechanisms responsible for postural orthostatic tachycardia syndrome (POTS) are incompletely understood. However, it is thought to be associated with absolute hypovolemia, hyperadrenergic responses, imbalance in the autonomic nervous system, and impaired regulation of peripheral vascular resistance. Hence, nonpharmacologic (eg., supplementation with water and salt or exercise) and pharmacologic (eg., fludrocortisones, midodrine hydrochloride [MD], beta blockers, pyridostigmine, or octreotide) therapies are applied in the management of POTS. However, the efficacy of most therapies is poor, and only a few agents have been reported to be effective and safe in adolescent patients. MD is used for treating POTS to improve impaired regulation of peripheral vascular resistance and venous pooling. Therefore, we hypothesized that children suffering from POTS with an excessive vasodilatory state might have better responses to MD. Flow-mediated vasodilation (FMD) by color Doppler vascular ultrasound is a noninvasive method of evaluating endothelial function. In our past research, our team revealed that children with POTS had greater FMD values than healthy children and suggested that augmented FMD and abnormal endothelial function might play important roles in the development of adolescent POTS. Therefore, the present study was undertaken to investigate whether FMD could be a predictor of therapeutic response to MD in children with POTS.
Methods
This prospective study was approved by the Ethics Committee of Peking University First Hospital (Beijing, China). All participants’ guardians were fully informed of the purpose and methods of our study, and informed consent was obtained.
Children who complained of orthostatic intolerance (OI) and were diagnosed as having POTS by the head-up test (HUT) or head-up tilt test (HUTT) between June 2008 to August 2011 at Peking University First Hospital were enrolled as POTS group. Healthy children without a history of OI and with a negative response to the HUT were selected as the control group.
The present study used the specific criteria to define POTS, as follows. (1) A child has a normal heart rate when supine and no evidence of cardiovascular disease, (2) after standing up or getting out of bed, a child has ≥3 of the following symptoms: dizziness, chest distress, chest pain, headache, palpitations, pale face, amaurosis, fatigue, discomfort, or syncope. These symptoms should be relieved or diminished by recumbence and should occur repeatedly for ≥1 month, (3) in addition to symptoms of OI, the child displays a heart rate increase ≥30 beats/min or a heart rate >120 beats/min within the first 10 minutes after standing or during HUT or HUTT. Simultaneously, the decrease in blood pressure should be <20/10 mm Hg, and (4) children with other diseases that manifest symptoms in the autonomic nervous system (eg., anemia, arrhythmia, hypertension, endocrine disorders) as well as cardiac or neurologic diseases that would induce syncope, were excluded.
The protocol for HUT or HUTT was according to the previously published literatures. Measurement of FMD was taken using a color Doppler ultrasound system (Ultrasound Cardiograph, HP2500, Philips Healthcare, Andover, Massachusetts), and the frequency of the transducer was 7.5 MHz. The protocol of measurement used in this study for endothelium-dependent FMD in the brachial artery was as recommended by the American College of Cardiology and other sources.
As preparation, participants were told to fast and avoid caffeine, high-fat foods, vitamin C, and vasoactive drugs with ≥4 half-lives. Exercise was also avoided for 4 to 6 hours before the examination. For female participants, information about menstrual cycles was requested, and the menstrual phase was avoided. All assessments were done between 8 a.m. and 9 a.m . The participant was supine, with his or her arm in a comfortable position for imaging of the brachial artery. The transducer was placed longitudinally on the skin of the upper arm approximately 5 to 10 cm above the antecubital fossa. The cuff of a mercury sphygmomanometer was placed around the forearm. A baseline image of the brachial artery was obtained to measure the diameter (baseline diameter) of the vascular lumen. The cuff was inflated to approximately 40 to 50 mm Hg higher than the systolic blood pressure to create an arterial occlusion. Inflation was maintained for 5 minutes. When the cuff was deflated, an image of the brachial artery was obtained again at the same position within 2 minutes after deflation of the cuff. The maximum diameter of the vascular lumen during the dilation was recorded. The FMD was calculated by using the formula (maximum diameter−baseline diameter)/baseline diameter × 100%.
All FMD measurements were done by the same operator. The stability and reproducibility of the detections were tested before the measurements. The data have been published previously. We also tested the reproducibility of FMD detection between operators. Two operators, including the operator in this study, each performed FMD detection on 10 volunteers. The values of FMD obtained by the 2 operators were compared.
Children with POTS who received MD (2.5 mg/day without other medications) for treatment were followed up for 3 months by telephone or by clinic visits. During follow-up, the symptoms and frequency of occurrence for each patient were recorded at 2 time points (1 month and 3 months) to evaluate the therapeutic effects of MD. For patients who came back to the clinic after 1 month, HUT and FMD of the brachial artery were evaluated again.
Symptom scoring was applied to evaluate the therapeutic effect of MD. Scorings were based mainly on the typical symptoms of OI, including syncope, dizziness or lightheadedness, chest tightness, nausea, palpitation, and headache. The score for each symptom was determined by its frequency according to the scoring system. If the child had never suffered from any of the aforementioned symptoms, he or she received a score of 0; if symptoms experienced less than once per month, a score of 1; if 2 to 4 times per week, a score of 2; if 2 to 7 times per week, a score of 3; and if more frequent than once per day, a score of 4. The total score was the sum of all symptom scores. Symptom scoring was performed for every participant at the beginning and after 1 month of therapy using MD. Therapy was thought to be effective if the symptom scores decreased by ≥2 points.
Statistical analyses were undertaken using SPSS ver. 13.0 (SPSS, Chicago, Illinois). Measurement data are the means ± SD. Numerative data are stated as cases. Comparisons between the POTS group and control group were performed using the independent t test for measurement data and chi-square test for numerative data. The paired samples correlation test and paired t test were used to evaluate the reproducibility of the FMD detection. The paired t test was used for comparisons of data before and after treatment. Comparisons between responders and nonresponders to MD were made by the independent t test. A p value <0.05 was considered significant.
A receiver operating characteristic (ROC) curve was applied to evaluate the value of FMD for predicting the therapeutic effect of MD. The area under the curve (AUC) indicated the predictive value of FMD. The notion that FMD is a predictor of the therapeutic effects of MD for treating children with POTS would be supported if p <0.05 and the 95% confidence interval (CI) of AUC was limited in (0 to 0.5) or (0.5 to 1). In detail, an AUC from 0.5 to 0.7 denoted a low predictive value; AUC from 0.7 to 0.9 represented a moderate predictive value, and AUC >0.9 reflected a high predictive value.
Results
In total, there were 108 children in POTS group and 20 children in the control group. Symptom scores of children in the control group were 0. Baseline data are shown in Table 1 .
| Characteristics | POTS Group (n = 108) | Control Group (n = 20) | t or Chi-Square Value | p Value |
|---|---|---|---|---|
| Male/female | 48/60 | 8/12 | 0.135 | 0.713 |
| Age (yrs) | 12 ± 3 | 12 ± 4 | −0.015 | 0.988 |
| Height (cm) | 151 ± 14 | 146 ± 20 | 1.021 | 0.318 |
| Body weight (kg) | 42 ± 13 | 46 ± 22 | −0.797 | 0.434 |
| Baseline diameter of brachial artery (mm) | 3.0 ± 0.4 | 3.1 ± 0.5 | −0.518 | 0.605 |
| FMD (%) | 11.0 ± 3.3 | 5.6 ± 2.2 | 7.057 | <0.001 |
| Mean artery pressure (mmHg) | 81 ± 8 | 79 ± 10 | 1.090 | 0.278 |
| Supine heart rate (beats/min) | 76 ± 10 | 77 ± 13 | −0.123 | 0.903 |
| Increased heart rate during HUT (beats/min) | 38 ± 9 | 7 ± 7 | 0.107 | <0.001 |
| Symptom score | 4.0 ± 2.2 | 0 | — | — |
The values for brachial artery baseline diameter and FMD detected by 2 operators were correlated, and there was no significant difference between them ( Table 2 ).
| Baseline Diameter of Brachial Artery (cm) | FMD (%) | |||
|---|---|---|---|---|
| Operator 1 | Operator 2 | Operator 1 | Operator 2 | |
| 1 | 2.9 | 3.2 | 10.34 | 8.57 |
| 2 | 3.5 | 3.2 | 5.71 | 5.88 |
| 3 | 3.5 | 3.6 | 8.57 | 7.69 |
| 4 | 3.3 | 3.4 | 9.09 | 8.11 |
| 5 | 3.8 | 3.8 | 5.26 | 5.00 |
| 6 | 3.8 | 3.7 | 5.26 | 7.50 |
| 7 | 3.0 | 3.2 | 6.67 | 5.88 |
| 8 | 3.1 | 3.1 | 6.45 | 8.82 |
| 9 | 4.2 | 4.0 | 2.33 | 2.44 |
| 10 | 4.1 | 4.1 | 4.88 | 4.88 |
| Paired samples correlation | r = 0.926 p <0.01 | r = 0.819 p <0.01 | ||
| Paired t test | p = 0.864 | p = 0.972 | ||
During treatment with MD, all 108 patients were rescored to evaluate the efficacy of MD after 1 month of therapy, whereas 105 were rescored after 3 months. Data for 3 participants were not available at the end point because the parents of the participants refused for their children to be followed up. Compared with baseline symptom scores (4.0 ± 2.2), higher symptom scores were reduced significantly after 1 of month treatment (1.0 ± 1.5; p <0.05) and 3 months of treatment (0.6 ± 1.3; p <0.05). HUT was performed and FMD was measured again for 68 patients who returned to clinic after 1 month of therapy. There was no difference in baseline characteristics (age, gender, height, weight, mean artery pressure, FMD, increased heart rates during HUT, symptom score) between the patients who had repeat HUT and those who did not (p >0.05, Supplementary Table 1 ). The increased FMD values and excessive heart rate during HUT were reduced significantly after treatment with MD (FMD: 11 ± 3% at baseline to 8 ± 3% after 1 month, p <0.05; delta heart rate: 38 ± 9 beats/min at baseline to 30 ± 11 beats/min after 1 month, p <0.05).
The data of symptom scores, changes in heart rates during HUT/HUTT, and FMD values before treatment were compared between the responders and nonresponders to MD at various time points during follow-up ( Table 3 ). There was a significant difference between responders and nonresponders with regard to FMD before treatment (p <0.05).
