Pediatr Cardiol DOI 10.1007/s00246-013-0843-9
ORIGINAL ARTICLE
Difference between Supine and Upright Blood Pressure Associates to the Efficacy of Midodrine on Postural Orthostatic Tachycardia Syndrome (POTS) in Children Wenjun Deng • Yanling Liu • Angie Dong Liu • Lukas Holmberg • Todd Ochs • Xueying Li • Jinyan Yang • Chaoshu Tang • Junbao Du Hongfang Jin
•
Received: 1 July 2013 / Accepted: 6 November 2013 Ó Springer Science+Business Media New York 2013
Abstract Postural orthostatic tachycardia syndrome (POTS) is common, and has a serious impact on children’s quality of life. Midodrine hydrochloride, an a1-adrenoreceptor agonist, is an effective treatment. The study was designed to examine the therapeutic efficacy of midodrine hydrochloride by quantifying changes in blood pressure during the head-up test (HUT), in children with POTS. Overall, 104 out of 110 children with POTS were treated with midodrine hydrochloride and successfully followedup. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) changes were analyzed during the HUT. In W. Deng J. Yang J. Du H. Jin (&) Department of Pediatrics, Peking University First Hospital, Beijing 100034, China e-mail: [email protected] Y. Liu Chaling Center for Disease Control and Prevention, Zhuzhou 412400, China A. D. Liu L. Holmberg Department of Medical and Health Sciences, Linko¨ping University, 58183 Linko¨ping, Sweden T. Ochs Feinberg School of Medicine, c/o The Ann and Robert Lurie Children’s Hospital, Chicago, IL 60611, USA X. Li Department of Medical Statistics, Peking University First Hospital, Beijing 100034, China C. Tang Department of Physiology and Pathophysiology, Peking University Health Sciences Center, Beijing 100191, China C. Tang Key Laboratory of Molecular Cardiology, Ministry of Education, Beijing, China
a retrospective analysis, a receiver operating characteristic (ROC) curve was used to analyze the therapeutic predictive value of pre-treatment changes in SBP, DBP, and a combination of both, from the supine position to standing, in the subjects. The increase of SBP and DBP from the supine position to standing in responders were significantly lower than that of the non-responders. The ROC curve showed that midodrine hydrochloride for children with POTS would be predicted to be effective when the pre-treatment increase of SBP was B0 mmHg, or when the pre-treatment increase of DBP was B6.5 mmHg (from the supine position to standing), yielding a sensitivity of 72 % and specificity of 88 %. The area under the curve was 0.744 and 0.809, respectively. Hence, the results suggested that looking at the changes in blood pressure during the HUT was useful in predicting the response to midodrine hydrochloride in children with POTS. Keywords Postural orthostatic tachycardia syndrome Midodrine hydrochloride Predictor Receiver operating characteristic curve
Introduction Postural orthostatic tachycardia syndrome (POTS) is a clinical syndrome presenting with orthostatic intolerance (OI) symptoms and a significant increase of the heart rate when standing [11, 12, 19]. In 1993, Schondorf and Low [16] described POTS by examining 16 patients who suffered from OI, and in 1999, Steward et al. [21] put forward the concept of POTS in childhood. Recently, POTS has become recognized as a major cause of OI in children [14], and has been shown to have a serious impact on the health and quality of life of children [1]. Therefore, the question
123
Pediatr Cardiol
of how to treat POTS effectively is important for pediatricians. Since the pathogenesis of POTS is very complex, the treatment modalities of POTS in children vary, and there are few randomized controlled treatment trials. Thus, in clinical practice it is often difficult to choose among the various therapeutic options for POTS. Generally speaking, the main treatments for POTS consist of non-drug therapies, including health education, training in upright positions, oral rehydration salts, and drug therapy, including bblockers, a-adrenoreceptor agonists, etc. [6]. In 2011, we found that midodrine hydrochloride was effective in treating POTS, with a cure rate of 68.42 % and an efficacy rate of 89.47 % [3]. However, a significant number of children did not respond to midodrine hydrochloride. Recently, our group discovered that plasma midregional pro-adrenomedullin (MR-proADM) of 61.5 % pg/ml produced both high sensitivity (100 %) and specificity (71.6 %) in predicting the efficacy of midodrine hydrochloride therapy for treating POTS in children [27]; however, the plasma measurement needs expenditure and is time-consuming. Therefore, we designed this study, attempting to predict in a simple and safe way which children with POTS were expected to have a favorable response to midodrine hydrochloride prior to initiating treatment. Hypovolemia, increased acetylcholine level in plasma, and abnormal vascular tension were documented as the main causes of POTS because each one affects blood pressure [8, 13]. Since an a1-adrenoreceptor agonist increases vascular tension, it was used to successfully treat some POTS patients who had reduced contractility of peripheral blood vessels and low peripheral vascular resistance [10]. Therefore, in clinical practice, it is important to select POTS children with a low vascular contractivity and resistance who might get a favorable response to midodrine hydrochloride. The head-up test (HUT) is a clinical test to induce a pathophysiologic status resembling that of POTS. Thus, is it possible that POTS cases having inadequate vascular resistance and blood pressure response during the HUT might be responsive to midodrine hydrochloride medication? Therefore, in the present study, a total of 104 children with POTS were studied to explore the possible value of changes in systolic blood pressure (SBP) and diastolic blood pressure (DBP) during the HUT, and examine its association with therapeutic efficacy of midodrine hydrochloride in children with POTS.
Materials and Methods Study Subjects and Grouping A total of 110 children, who were diagnosed with POTS and who were only treated with midodrine hydrochloride in
123
our hospital from 2005 to 2011, were enrolled in the study. The chronic symptoms of the children enrolled in the study included OI symptoms such as dizziness, syncope, nausea, palpitations, chest tightness, headache, blurred vision, tremors, cold sweat, etc., for at least 1 month [15, 27]. A total of 49 males, accounting for 44.55 %, and 61 females, accounting for 55.45 %, were included. Their ages ranged from 6 to 18 years, and the average age was 11.92 ± 2.51 years. Six children, who did not complete follow-up, were excluded. The remaining 104 patients with POTS were divided into two groups according to their response to the midodrine hydrochloride treatment: effective group (79 cases) and non-effective group (25 cases). Diagnostic Criterion Diagnostic criteria for diagnosing POTS in children were as follows [4, 18, 24]: (1) chronic OI symptoms, including dizziness, syncope, nausea, palpitations, chest tightness, headache, blurred vision, tremors, cold sweat, etc.; (2) the HUT was positive for POTS. Children with other diseases such as anemia, hypertension, cardiac arrhythmias, endocrine diseases, metabolic diseases, nervous system diseases, which can cause similar symptoms, were excluded from the study. Children who met all the above-mentioned criteria were diagnosed with POTS. Head-Up Test [2, 4, 18] For the HUT, children lay quietly in the supine position for 10 min, and the basal heart rate and blood pressure were recorded, and a routine ECG was obtained. The children were then asked to stand, and the heart rate, blood pressure, and ECG were monitored. Children were closely observed for any OI symptoms during the test. The test was considered positive when a child’s heart rate increased 30 beats/min or more, or if his/her heart rate reached at least 120 beats/min within the first 10 min of the HUT. In addition, any symptoms such as dizziness, vertigo, headache, palpitations, fatigue, or blurred vision were viewed as a positive test. Treatment and Follow-Up Before treatment, age, gender, basal supine and standing SBP, the change of SBP from the supine position to standing were recorded and calculated in children who were diagnosed with POTS. Midodrine hydrochloride treatment was then initiated in all patients, at 2.5 mg once a day, without other medications. Patients were followed-up for 6 months after treatment, by telephone calls or outpatient visits, and any changes in clinical symptoms were
Pediatr Cardiol
recorded. Six children who did not complete the follow-up were excluded from the study. Evaluation of Drug Effect [23] Depending on the frequency of clinical symptoms, including syncope, dizziness, chest tightness, palpitations, headache, blurred vision, trembling and cold sweat, the severity scores ranged from 0 to 4 points. The score was calculated as follows: 0 point indicated that clinical symptoms never appeared; 1 point that the frequency of clinical symptoms occurred on average once a month; 2 points that the frequency of clinical symptoms occurred on average 2–4 times a month; 3 points that the frequency of clinical symptoms occurred on average 2–7 times a week; and 4 points that the frequency of clinical symptoms occurred on average more than once a day. The basal clinical symptom score of each child was evaluated and recorded before treatment, and the clinical symptoms score was then evaluated and recorded during the follow-up. Treatment was deemed effective if the final clinical symptom score decreased by 2 points or more compared with the basal clinical symptom score. On the contrary, it was deemed ineffective if the final clinical symptoms score decreased by less than 2 points compared with the basal clinical symptom score. The differences in blood pressure (changes) in the supine and standing positions before the start of drug therapy of responders/non-responders were retrospectively analysed. Statistical Analysis The collected data were expressed by mean ± standard error. SPSS 14.0 was used for all statistical analyses. The single Kolmogorov–Smirnov test was used to verify whether the pre-treatment blood pressure data were normally distributed, and the independent sample t test was used to compare the pre-treatment blood pressure data between the effective group and the ineffective group. Otherwise, the pre-treatment blood pressure data were compared between the two groups by the rank sum test. The receiver operating characteristic (ROC) curve was used to explore the predictive value of the changes in blood pressure during the HUT to predict therapeutic efficacy of midodrine hydrochloride in children with POTS. Additionally, the ROC curve allowed the use of parallel association and series connection of the changes in pre-treatment blood pressure during the HUT to predict therapeutic efficacy of midodrine hydrochloride in these children. If the 95 % confidence interval (CI) of the area under the curve did not contain 0.5, or the p value was \0.05, the test index was considered to have predictive value. The v2 test was used to
compare the efficacy rates between children with different SBP or DBP changes, depending upon the cut-off value derived from ROC analysis. Statistical significance was defined as p \ 0.05.
Results Clinical Data Comparison Between the Effective and Ineffective Groups There were no statistical differences between the effective group and the ineffective group in age, duration of symptoms, heart rate before or after the HUT, changes in heart rate during the HUT, supine SBP and DBP, symptom scores before treatment, and standing SBP before treatment during the HUT (p [ 0.05). The standing DBP, and the changes in SBP and DBP from the supine position to standing, before treatment during the HUT in the effective group were lower than those of the ineffective group (standing DBP: 67.4 ± 8.7 vs. 72.6 ± 7.4 mmHg, p \ 0.05; delta SBP: 0.3 ± 6.9 vs. 6.2 ± 10.7 mmHg, p \ 0.01; and delta DBP: 3.8 ± 8.5 vs. 12.9 ± 7.0 mmHg, p \ 0.01; Table 1). Association of the Therapeutic Efficacy of Midodrine Hydrochloride with Blood Pressure Changes in Children with POTS from the Supine Position to Standing, Utilizing Receiver Operating Characteristic Curve Analysis. Before treatment during the HUT, an ROC curve analysis was conducted in order to explore the suitability of using SBP changes from the supine position to standing to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.744 (95 % CI 0.616–0.871, p \ 0.05) and the cut-off value was 0 mmHg. The therapeutic efficacy of midodrine hydrochloride in children with POTS was hypothesized to be effective when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg. Predictive sensitivity was 60 % and predictive specificity was 80 % (Fig. 1). An ROC curve analysis was also conducted to explore the probability of using DBP changes from the supine position to standing before treatment during the HUT to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.809 (95 % CI 0.718–0.899, p \ 0.05), and the cut-off value was 6.5 mmHg. We confirmed that midodrine hydrochloride in children with POTS would be effective when the increase in DBP from the supine position to standing during the HUT before treatment was B6.5 mmHg. Predictive sensitivity was 63 % and predictive specificity was 92 % (Fig. 2).
123
Pediatr Cardiol Table 1 The age and pre-treatment hemodynamics during the headup test of both the effective and the ineffective groups Items
Effective group (n = 79)
Ineffective group (n = 25)
t, z or f value
p value
t = 0.32
0.75
t = -1.01
0.32
t = 0.81
0.42
Age (years)
11.8 ± 2.6
12.0 ± 2.4
DOS (mon)
6.5 ± 2.7
7.2 ± 3.7
SHR (bpm)
75.3 ± 10.3
73.3 ± 12.3
UHR (bpm)
108.9 ± 13.6
108.9 ± 13.6
t = -0.22
0.82
33.6 ± 11.6
34.9 ± 7.8
z = -0.03
0.98
Delta HR (bpm) SSBP (mmHg)
103.7 ± 9.3
102.4 ± 11.3
t = -0.54
0.59
USBP (mmHg)
104.0 ± 10.1
108.7 ± 12.4
t = 1.91
0.06
6.2 ± 10.7
z = -3.70
0.00*
Delta SBP (mmHg)
0.3 ± 6.9**
SDBP (mmHg)
63.6 ± 9.4
59.7 ± 10.0
t = -1.79
0.08
UDBP (mmHg)
67.4 ± 8.7*
72.6 ± 7.4
t = 2.70
0.01*
12.9 ± 7.0
z = -4.69
0.00*
5.0 ± 0.8
t = -0.53
0.595
4.7 ± 0.9
t = -15.6
0.00*
Delta DBP (mmHg)
3.8 ± 8.5**
SSBT
4.8 ± 0.9
SSAT
4
1.3 ± 0.9**
DOS duration of the symptoms, SHR supine heart rate, UHR upright heart rate, Delta HR the heart rate difference from the supine position to upright, SSBP supine systolic blood pressure, USBP upright systolic blood pressure, Delta SBP systolic blood pressure difference from the supine position to upright, SDBP supine diastolic blood pressure, UDBP upright diastolic blood pressure, Delta DBP diastolic blood pressure difference from the supine position to upright, SSBT symptom score before treatment, SSAT symptom score after treatment * p \ 0.05 versus ineffective group; ** p \ 0.01 versus ineffective group, p \ 0.01: compare symptom score between before treatment and after treatment in the effective group
To accurately predict the therapeutic efficacy of midodrine hydrochloride, an ROC curve analysis of the parallel association of SBP and DBP changes during the HUT, a model of combined analysis, was utilized. The results confirmed that midodrine hydrochloride was expected to be effective in children with POTS when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg, or when the increase in DBP from the supine position to standing during the HUT before treatment was B6.5 mmHg. Predictive sensitivity was 72 % and specificity was 88 %. In addition, ROC curve series connection of SBP and DBP changes, from the supine position to standing, revealed that the therapeutic efficacy of midodrine
123
Fig. 1 The ROC curve analysis based on the systolic pressure difference from the supine position to standing and the therapeutic efficacy of midodrine hydrochloride in children with POTS. Before treatment during the HUT, an ROC curve analysis was conducted in order to explore the suitability of using SBP changes from the supine position to standing to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.744 (95 % CI 0.616–0.871; p \ 0.05) and the cut-off value was 0 mmHg. The therapeutic efficacy of midodrine hydrochloride in children with POTS was hypothesized to be effective, when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg. The predictive sensitivity was 60 % and the predictive specificity was 80 %. The sensitivity is expressed on the longitudinal axis and 1-specificity on the horizontal axis. ROC receiver operating characteristic, POTS postural orthostatic tachycardia syndrome, HUT head-up test, SBP systolic blood pressure
hydrochloride in children with POTS was effective when the increase in SBP from the supine position to standing before treatment was B0 mmHg, and the increase in DBP from the supine position to standing before treatment was B6.5 mmHg. Predictive sensitivity was 58 % and specificity was 92 %. Comparison of the Effective Rate Based on the Cut-Off Values of Blood Pressure Changes During the Head-Up Test in Children with POTS. At the 6-month follow-up, the efficacy of midodrine hydrochloride in patients with an increase in SBP B 0 mmHg from the supine position to standing before treatment was significantly higher than those with an increase in SBP over 0 mmHg from the supine position to standing (90.3 vs. 61.5 %, p \ 0.01; Table 2). Similarly, the efficacy rate for midodrine hydrochloride in patients with an increase in DBP B 6.5 mmHg from the supine
Pediatr Cardiol Table 2 Comparison of the effective rate based on the cut-off values of SBP changes during the HUT in children with POTS Pre-treatment SBP increase from spine to upright
Effective cases (n)
Ineffective cases (n)
Total number (n)
Effective rate (%)
[0 mmHg
32
20
52
61.5
^0 mmHg
47
5
52
90.3*
SBP systolic blood pressure, HUT head-up test, POTS postural orthostatic tachycardia syndrome * p \ 0.01 versus [0 mmHg
Table 3 Comparison of the effective rate based on the cut-off values of DBP changes during the HUT in children with POTS Pre-treatment DBP increase from spine to upright
Fig. 2 The ROC curve analysis was based on the diastolic pressure difference from the supine position to standing and the therapeutic efficacy of midodrine hydrochloride in children with POTS. A ROC curve analysis was also conducted to explore the probability of using DBP changes from the supine position to standing before treatment during the HUT to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.809 (95 % CI 0.718–0.899; p \ 0.05) and the cut-off value was 6.5 mmHg. We confirmed that midodrine hydrochloride in children with POTS would be effective when the increase in DBP from the supine position to standing during HUT before treatment was B6.5 mmHg. The predictive sensitivity was 63 % and the predictive specificity was 92 %. The sensitivity is expressed on the longitudinal axis and 1-specificity on the horizontal axis. ROC receiver operating characteristic, POTS postural orthostatic tachycardia syndrome, DBP diastolic blood pressure, HUT head-up test
position to standing before treatment was significantly higher than those with an increase in DBP over 6.5 mmHg from the supine position to standing (95.0 vs. 64.1 %, p \ 0.01; Table 3).
Discussion In his study of nine patients treated with midodrine hydrochloride, Hoeldtke et al. [7] found that midodrine hydrochloride improved abnormal changes of the heart rate of patients with POTS during the HUT. Grubb et al. [5] believed that midodrine hydrochloride partially improved symptoms of OI. We previously compared the effective rate of different treatment modalities (midodrine hydrochloride plus oral rehydration salts, metoprolol plus oral rehydration salts, and oral rehydration salts only) and found that midodrine hydrochloride plus oral rehydration salts
Effective cases (n)
Ineffective cases (n)
Total number (n)
Effective rate (%)
[6.5 mmHg
41
23
64
64.1
^6.5 mmHg
38
2
40
95.0*
DBP diastolic blood pressure, HUT head-up test, POTS postural orthostatic tachycardia syndrome * p \ 0.01 versus [6.5 mmHg
were effective in some children with POTS [26]. In 2011, we studied 53 children with POTS treated with midodrine hydrochloride, and confirmed that there were some children with POTS who were unresponsive to midodrine hydrochloride [3]. Therefore, it is essential to be able to predict the efficacy of midodrine hydrochloride in children with POTS prior to initiating such therapy. The dose of midodrine hydrochloride used to treat POTS effectively is controversial, and there have been few studies indicating the standard dose of midodrine in pediatric medicine. In this study, the midodrine dose was 2.5 mg/day in accordance with previously published papers regarding pediatric cases [3, 9, 20, 25]. However, this dose seems relatively low and because of its short-acting nature, increasing the number of doses per day should be considered. In the future, additional studies are needed to clarify the use and dosage of midodrine in the pediatric population based on the fact that it is a short-acting drug. How to evaluate the therapeutic effect of midodrine hydrochloride on POTS was a difficult issue. There were several evaluating criteria, including the HUT outcome and symptoms, etc. As one of the criteria, the symptom score based on the typical symptoms of OI has been used; it is practical and simple and can quantify the therapeutic effect [27]. Therefore, in this study, we used this criterion to evaluate the therapeutic effect of midodrine hydrochloride in children with POTS, although using the symptom score together with symptom severity and repeated HUT would be more accurate.
123
Pediatr Cardiol
In this study, we discovered that the increase of standing SBP and DBP compared with the supine position before treatment in the effective group was markedly less than that of the ineffective group. Furthermore, ROC analysis showed that midodrine hydrochloride in children with POTS was expected to be effective when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg, or the increase in DBP from the supine position to standing during the HUT before treatment was B6.5 mmHg. Furthermore, the predictive sensitivity and specificity increased when using a combined modality parallel association ROC analysis of standing SBP change from the supine position, or standing DBP change from the supine position. However, predictive sensitivity for the series connection ROC analysis of standing SBP changes from the supine position to standing, and the DBP change from the supine position, was not favored, but specificity was high. Additionally, we found that the efficacy rate of midodrine hydrochloride in patients with increases in SBP or DBP below the cut-off values (from the supine position to standing, before treatment) was significantly higher than those with increases in SBP or DBP beyond the cut-off values. Based upon the above results, we reported that the therapeutic efficacy of midodrine hydrochloride in children with POTS could be predicted by measuring blood pressure during the HUT before treatment. Patients would get a better therapeutic response when the increase in SBP from the supine position to standing was B0 mmHg, or the increase in DBP from the supine position to standing was B6.5 mmHg. This information could greatly aid in the process of choosing which children with POTS would be responsive to midodrine hydrochloride. Recently, it was discovered that plasma MR-proADM of 61.5 % pg/ml produced both high sensitivity (100 %) and specificity (71.6 %) in predicting the efficacy of midodrine hydrochloride therapy for treating POTS in children [27]. Compared with plasma MR-proADM, measuring blood pressure had lower sensitivity (72 vs. 100 %) but higher specificity (88 vs. 71.6 %) in the prediction. At the same time, measuring blood pressure is simple, cheap, and noninvasive compared with the detection of MR-proADM levels. Therefore, using the SBP or DBP changes during the HUT before treatment to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS is feasible and cost effective in clinical practice. The underlying mechanisms by which a relatively small increase or decrease or no change in blood pressure during the HUT predicts a better response to midodrine hydrochloride treatment in children with POTS, remain unclear. They might be associated with the pathogenesis of POTS and the consequent pharmacological effects of midodrine hydrochloride. The pathogenesis of POTS is complex and
123
presently unclear, but low blood volume, abnormal local blood flow, and autonomic stabilization of vascular tension are all hypothetical mechanisms [6, 13]. When children with POTS change position from the supine position to standing, their venous return is reduced, decreasing blood supply for the major organs, especially the brain, causing the OI symptoms to appear. The a1-adrenoreceptor agonist, midodrine hydrochloride, can increase the contractility of arterial vessels, maintaining the blood supply to the brain and thereby improving OI symptoms [17, 22]. Since the decreased venous return lowers the blood pressure, changes in blood pressure during the HUT seem useful in predicting the response to midodrine hydrochloride in children with POTS. The majority of cases with a relatively small increase or decrease or without any change in standing blood pressure versus supine blood pressure during the HUT test might be reflective of the suitability of the use of a1-adrenoreceptor agonists for children with POTS. However, although in each group some children responded well to the lowering of blood pressure during HUT, some did not. The mechanisms might involve the facts that midodrine hydrochloride as an a1-adrenoreceptor agonist functions via various pathways, including regulating the inappropriate control of peripheral vascular resistance that has been supposed to be a cause of POTS [12], decreasing sympathetic nervous activity and improving abnormal flow-mediated dilation in children with POTS [10]. In the study, the negative predictive values for SDP and DBP based on the predictive criterion were 38.5 and 36.0 %; namely, 61.5 and 64 % of patients who render the cut-off of the SDP and DBP, respectively, still respond to midodrine hydrochloride. The positive predictive values for SDP and DBP based on the predictive criterion were 90.3 and 95 %; namely, almost all patients who accord with the cut-off of the SDP and DBP, respectively, ideally respond to midodrine hydrochloride. Based on the above results, we suggested that children with POTS who accorded with the predictive criterion should, as the first choice, be treated with midodrine hydrochloride, while for those who render the predictive criterion, midodrine hydrochloride may be an alternative rather than the first choice. However, this present study has several limitations. Ideally, the inclusion of a placebo POTS group, where the patients’ outcome and prediction of BP change from the supine position to standing should have been analyzed and compared with the results of the present studies, would be meaningful for excluding the possibility of a natural history of symptoms with no relation to the midodrine. However, it was impractical to treat the children with POTS with placebo, without medication. Further studies including a placebo group are necessary. In this study, we only followedup the patients for 6 months, and were unable to know how
Pediatr Cardiol
long the therapeutic effect would last after the study period. Therefore, in the future, a long-term follow-up study of the value of hemodynamics in predicting the efficacy of midodrine hydrochloride in children with POTS is needed. In addition, a prospective cohort study for these criteria found in the present study should be designed in the future to see the generalizabilty of the findings. Acknowledgments This work was supported by the Major Basic Research Project of China (2012CB517806) and the National Twelfth Five-Year Plan for Science & Technology Support (2012BAI03B03). We thank Dr Jing Lin for her work on data processing. Conflict of interest
None.
References 1. Benrud-Larson LM, Dewar MS, Sandroni P, Rummans TA, Haythornthwaite JA, Low PA (2002) Quality of life in patients with postural tachycardia syndrome. Mayo Clin Proc 77:531–537 2. Chen L, Yang YY, Wang C, Wang HW, Tian H, Zhang QY, Chen JJ, Wang YL, Kang YL, Tang CS, Du JB (2006) A multicenter study of hemodynamic characteristics exhibited by children with unexplained syncope. Chin Med J 119:2062–2068 3. Chen L, Wang L, Sun J, Qin J, Tang C, Jin H, Du J (2011) Midodrine hydrochloride is effective in the treatment of children with postural orthostatic tachycardia syndrome. Circ J 75: 927–931 4. Chinese Pediatric Cardiology Society, Editorial Board of Chinese Journal of Pediatrics (2009) Guideline of diagnosis of syncope in children. Chin J Pediatr 47:99–100 5. Grubb BP, Karas B, Kosinski D, Boehm K (1999) Preliminary observations on the use of midodrine hydrochloride in the treatment of refractory neurocardiogenic syncope. J Interv Card Electrophysiol 3:139–143 6. Grubb BP, Kanjwal MY, Kosinski DJ (2001) The postural orthostatic tachycardia syndrome: current concepts in pathophysiology diagnosis and management. J Interv Card Electrophysiol 5(1):9–14 7. Hoeldtke RD, Bryner KD, Hoeldtke ME, Hobbs G (2006) Treatment of postural tachycardia syndrome: a comparison of octreotide and midodrine. Clin Auton Res 16:390–395 8. Julian MS, Amy W (2000) Vascular perturbations in the chronic orthostatic intolerance of the postural orthostatic tachycardia syndrome. J Appl Physiol 89:1505–1512 9. Kaufmann H, Saadia D, Voustianiouk A (2002) Midodrine in neurally mediated syncope: a double-blind randomized crossover study. Ann Neurol 52:342–345
10. Liao Y, Yang J, Zhang F et al (2013) Flow-mediated vasodilation as a predictor of therapeutic response to midodrine hydrochloride in children with postural orthostatic tachycardia syndrome. Am J Cardiol 112(6):816–820 11. Low PA, Sandroni P, Joyner M, Shen WK (2009) Postural tachycardia syndrome (POTS). J Cardiovasc Electrophysiol 20:352–358 12. Medow MS, Stewart JM (2007) The postural tachycardia syndrome. Cardiol Rev 15:67–75 13. Raj SR, Biaggioni I, Yamhure PC, Black BK, Paranjape SY, Byrne DW, Robertson D (2005) Renin-aldosterone paradox and perturbed blood volume regulation underlying postural tachycardia syndrome. Circulation 111:1574–1582 14. Rorbertson D (1999) The epidemic of orthostatic tachycardia and orthostatic intolerance. Am J Med Sci 317:75–77 15. Sandroni P, Tonette L, Benjamin R, Phillip AL (1999) Postural tachycardia syndrome: clinic features and follow-up study. Mayo Clin Proc 74:1106–1110 16. Schondorf R, Low PA (1993) Idiopathic postural orthostatic tachycardia syndrome: an attenuated form of acute pandysautonomia? Neurology 43(1):132–137 17. Sewarts JA (1988) Measuring the accuracy of diagnostic systems. Science 240:1285–1293 18. Steward JM (2002) Orthostatic intolerance in pediatrics. J Pediatr 140:404–450 19. Stewart JM (2004) Chronic orthostatic intolerance and the postural tachycardia syndrome (POTS). J Pediatr 145:725–730 20. Stewart JM (2006) Midodrine for the treatment of vasovagal syncope (simple faint). J Pediatr 149:740–742 21. Stewart JM, Gewitz MH, Weldon A, Munoz J (1999) Patterns of orthostatic intolerance: the orthostatic tachycardia syndrome and adolescent chronic fatigue. J Pediatr 135:218–225 22. Vaddadi G, Lambert E, Corcoran SJ, Esler MD (2007) Postural syncope: mechanisms and management. Med J Aust 187:299–304 23. Winder R, Barth A, Bidmon D, Ponocny I, Weber M, Mayr O, Robertson D, Diedrich A, Maier R, Pilger A, Haber P, Ru¨diger HW (2005) Endurance exercise training in orthostatic intolerance: a randomized controlled trial. Hypertension 45(3):391–398 24. Zhang QY, Du JB, Li WZ (2005) Clinical analysis and follow-up study of postural orthostatic tachycardia syndrome in 28 pediatric cases. Chin J Pediatr 43:165–169 25. Zhang Q, Du J, Tang C (2006) The efficacy of midodrine hydrochloride in the treatment of children with vasovagal syncope. J Pediatr 149:777–780 26. Zhang FW, Liao Y, Li XY, Chen L, Jin HF, Du JB (2011) An analysis of 118 cases for midodrine hydrochloride and metoprolol treating postural tachycardia syndrome in children. Chin J Pediatr 49:428–432 27. Zhang F, Li X, Ochs T, Chen L, Liao Y, Tang C, Jin H, Du J (2012) Midregional pro-adrenomedullin as predictor for therapeutic response to midodrinehydrochloride in children with postural orthostatic tachycardia syndrome. J Am Coll Cardiol 60:315–320
123
ORIGINAL ARTICLE
Difference between Supine and Upright Blood Pressure Associates to the Efficacy of Midodrine on Postural Orthostatic Tachycardia Syndrome (POTS) in Children Wenjun Deng • Yanling Liu • Angie Dong Liu • Lukas Holmberg • Todd Ochs • Xueying Li • Jinyan Yang • Chaoshu Tang • Junbao Du Hongfang Jin
•
Received: 1 July 2013 / Accepted: 6 November 2013 Ó Springer Science+Business Media New York 2013
Abstract Postural orthostatic tachycardia syndrome (POTS) is common, and has a serious impact on children’s quality of life. Midodrine hydrochloride, an a1-adrenoreceptor agonist, is an effective treatment. The study was designed to examine the therapeutic efficacy of midodrine hydrochloride by quantifying changes in blood pressure during the head-up test (HUT), in children with POTS. Overall, 104 out of 110 children with POTS were treated with midodrine hydrochloride and successfully followedup. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) changes were analyzed during the HUT. In W. Deng J. Yang J. Du H. Jin (&) Department of Pediatrics, Peking University First Hospital, Beijing 100034, China e-mail: [email protected] Y. Liu Chaling Center for Disease Control and Prevention, Zhuzhou 412400, China A. D. Liu L. Holmberg Department of Medical and Health Sciences, Linko¨ping University, 58183 Linko¨ping, Sweden T. Ochs Feinberg School of Medicine, c/o The Ann and Robert Lurie Children’s Hospital, Chicago, IL 60611, USA X. Li Department of Medical Statistics, Peking University First Hospital, Beijing 100034, China C. Tang Department of Physiology and Pathophysiology, Peking University Health Sciences Center, Beijing 100191, China C. Tang Key Laboratory of Molecular Cardiology, Ministry of Education, Beijing, China
a retrospective analysis, a receiver operating characteristic (ROC) curve was used to analyze the therapeutic predictive value of pre-treatment changes in SBP, DBP, and a combination of both, from the supine position to standing, in the subjects. The increase of SBP and DBP from the supine position to standing in responders were significantly lower than that of the non-responders. The ROC curve showed that midodrine hydrochloride for children with POTS would be predicted to be effective when the pre-treatment increase of SBP was B0 mmHg, or when the pre-treatment increase of DBP was B6.5 mmHg (from the supine position to standing), yielding a sensitivity of 72 % and specificity of 88 %. The area under the curve was 0.744 and 0.809, respectively. Hence, the results suggested that looking at the changes in blood pressure during the HUT was useful in predicting the response to midodrine hydrochloride in children with POTS. Keywords Postural orthostatic tachycardia syndrome Midodrine hydrochloride Predictor Receiver operating characteristic curve
Introduction Postural orthostatic tachycardia syndrome (POTS) is a clinical syndrome presenting with orthostatic intolerance (OI) symptoms and a significant increase of the heart rate when standing [11, 12, 19]. In 1993, Schondorf and Low [16] described POTS by examining 16 patients who suffered from OI, and in 1999, Steward et al. [21] put forward the concept of POTS in childhood. Recently, POTS has become recognized as a major cause of OI in children [14], and has been shown to have a serious impact on the health and quality of life of children [1]. Therefore, the question
123
Pediatr Cardiol
of how to treat POTS effectively is important for pediatricians. Since the pathogenesis of POTS is very complex, the treatment modalities of POTS in children vary, and there are few randomized controlled treatment trials. Thus, in clinical practice it is often difficult to choose among the various therapeutic options for POTS. Generally speaking, the main treatments for POTS consist of non-drug therapies, including health education, training in upright positions, oral rehydration salts, and drug therapy, including bblockers, a-adrenoreceptor agonists, etc. [6]. In 2011, we found that midodrine hydrochloride was effective in treating POTS, with a cure rate of 68.42 % and an efficacy rate of 89.47 % [3]. However, a significant number of children did not respond to midodrine hydrochloride. Recently, our group discovered that plasma midregional pro-adrenomedullin (MR-proADM) of 61.5 % pg/ml produced both high sensitivity (100 %) and specificity (71.6 %) in predicting the efficacy of midodrine hydrochloride therapy for treating POTS in children [27]; however, the plasma measurement needs expenditure and is time-consuming. Therefore, we designed this study, attempting to predict in a simple and safe way which children with POTS were expected to have a favorable response to midodrine hydrochloride prior to initiating treatment. Hypovolemia, increased acetylcholine level in plasma, and abnormal vascular tension were documented as the main causes of POTS because each one affects blood pressure [8, 13]. Since an a1-adrenoreceptor agonist increases vascular tension, it was used to successfully treat some POTS patients who had reduced contractility of peripheral blood vessels and low peripheral vascular resistance [10]. Therefore, in clinical practice, it is important to select POTS children with a low vascular contractivity and resistance who might get a favorable response to midodrine hydrochloride. The head-up test (HUT) is a clinical test to induce a pathophysiologic status resembling that of POTS. Thus, is it possible that POTS cases having inadequate vascular resistance and blood pressure response during the HUT might be responsive to midodrine hydrochloride medication? Therefore, in the present study, a total of 104 children with POTS were studied to explore the possible value of changes in systolic blood pressure (SBP) and diastolic blood pressure (DBP) during the HUT, and examine its association with therapeutic efficacy of midodrine hydrochloride in children with POTS.
Materials and Methods Study Subjects and Grouping A total of 110 children, who were diagnosed with POTS and who were only treated with midodrine hydrochloride in
123
our hospital from 2005 to 2011, were enrolled in the study. The chronic symptoms of the children enrolled in the study included OI symptoms such as dizziness, syncope, nausea, palpitations, chest tightness, headache, blurred vision, tremors, cold sweat, etc., for at least 1 month [15, 27]. A total of 49 males, accounting for 44.55 %, and 61 females, accounting for 55.45 %, were included. Their ages ranged from 6 to 18 years, and the average age was 11.92 ± 2.51 years. Six children, who did not complete follow-up, were excluded. The remaining 104 patients with POTS were divided into two groups according to their response to the midodrine hydrochloride treatment: effective group (79 cases) and non-effective group (25 cases). Diagnostic Criterion Diagnostic criteria for diagnosing POTS in children were as follows [4, 18, 24]: (1) chronic OI symptoms, including dizziness, syncope, nausea, palpitations, chest tightness, headache, blurred vision, tremors, cold sweat, etc.; (2) the HUT was positive for POTS. Children with other diseases such as anemia, hypertension, cardiac arrhythmias, endocrine diseases, metabolic diseases, nervous system diseases, which can cause similar symptoms, were excluded from the study. Children who met all the above-mentioned criteria were diagnosed with POTS. Head-Up Test [2, 4, 18] For the HUT, children lay quietly in the supine position for 10 min, and the basal heart rate and blood pressure were recorded, and a routine ECG was obtained. The children were then asked to stand, and the heart rate, blood pressure, and ECG were monitored. Children were closely observed for any OI symptoms during the test. The test was considered positive when a child’s heart rate increased 30 beats/min or more, or if his/her heart rate reached at least 120 beats/min within the first 10 min of the HUT. In addition, any symptoms such as dizziness, vertigo, headache, palpitations, fatigue, or blurred vision were viewed as a positive test. Treatment and Follow-Up Before treatment, age, gender, basal supine and standing SBP, the change of SBP from the supine position to standing were recorded and calculated in children who were diagnosed with POTS. Midodrine hydrochloride treatment was then initiated in all patients, at 2.5 mg once a day, without other medications. Patients were followed-up for 6 months after treatment, by telephone calls or outpatient visits, and any changes in clinical symptoms were
Pediatr Cardiol
recorded. Six children who did not complete the follow-up were excluded from the study. Evaluation of Drug Effect [23] Depending on the frequency of clinical symptoms, including syncope, dizziness, chest tightness, palpitations, headache, blurred vision, trembling and cold sweat, the severity scores ranged from 0 to 4 points. The score was calculated as follows: 0 point indicated that clinical symptoms never appeared; 1 point that the frequency of clinical symptoms occurred on average once a month; 2 points that the frequency of clinical symptoms occurred on average 2–4 times a month; 3 points that the frequency of clinical symptoms occurred on average 2–7 times a week; and 4 points that the frequency of clinical symptoms occurred on average more than once a day. The basal clinical symptom score of each child was evaluated and recorded before treatment, and the clinical symptoms score was then evaluated and recorded during the follow-up. Treatment was deemed effective if the final clinical symptom score decreased by 2 points or more compared with the basal clinical symptom score. On the contrary, it was deemed ineffective if the final clinical symptoms score decreased by less than 2 points compared with the basal clinical symptom score. The differences in blood pressure (changes) in the supine and standing positions before the start of drug therapy of responders/non-responders were retrospectively analysed. Statistical Analysis The collected data were expressed by mean ± standard error. SPSS 14.0 was used for all statistical analyses. The single Kolmogorov–Smirnov test was used to verify whether the pre-treatment blood pressure data were normally distributed, and the independent sample t test was used to compare the pre-treatment blood pressure data between the effective group and the ineffective group. Otherwise, the pre-treatment blood pressure data were compared between the two groups by the rank sum test. The receiver operating characteristic (ROC) curve was used to explore the predictive value of the changes in blood pressure during the HUT to predict therapeutic efficacy of midodrine hydrochloride in children with POTS. Additionally, the ROC curve allowed the use of parallel association and series connection of the changes in pre-treatment blood pressure during the HUT to predict therapeutic efficacy of midodrine hydrochloride in these children. If the 95 % confidence interval (CI) of the area under the curve did not contain 0.5, or the p value was \0.05, the test index was considered to have predictive value. The v2 test was used to
compare the efficacy rates between children with different SBP or DBP changes, depending upon the cut-off value derived from ROC analysis. Statistical significance was defined as p \ 0.05.
Results Clinical Data Comparison Between the Effective and Ineffective Groups There were no statistical differences between the effective group and the ineffective group in age, duration of symptoms, heart rate before or after the HUT, changes in heart rate during the HUT, supine SBP and DBP, symptom scores before treatment, and standing SBP before treatment during the HUT (p [ 0.05). The standing DBP, and the changes in SBP and DBP from the supine position to standing, before treatment during the HUT in the effective group were lower than those of the ineffective group (standing DBP: 67.4 ± 8.7 vs. 72.6 ± 7.4 mmHg, p \ 0.05; delta SBP: 0.3 ± 6.9 vs. 6.2 ± 10.7 mmHg, p \ 0.01; and delta DBP: 3.8 ± 8.5 vs. 12.9 ± 7.0 mmHg, p \ 0.01; Table 1). Association of the Therapeutic Efficacy of Midodrine Hydrochloride with Blood Pressure Changes in Children with POTS from the Supine Position to Standing, Utilizing Receiver Operating Characteristic Curve Analysis. Before treatment during the HUT, an ROC curve analysis was conducted in order to explore the suitability of using SBP changes from the supine position to standing to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.744 (95 % CI 0.616–0.871, p \ 0.05) and the cut-off value was 0 mmHg. The therapeutic efficacy of midodrine hydrochloride in children with POTS was hypothesized to be effective when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg. Predictive sensitivity was 60 % and predictive specificity was 80 % (Fig. 1). An ROC curve analysis was also conducted to explore the probability of using DBP changes from the supine position to standing before treatment during the HUT to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.809 (95 % CI 0.718–0.899, p \ 0.05), and the cut-off value was 6.5 mmHg. We confirmed that midodrine hydrochloride in children with POTS would be effective when the increase in DBP from the supine position to standing during the HUT before treatment was B6.5 mmHg. Predictive sensitivity was 63 % and predictive specificity was 92 % (Fig. 2).
123
Pediatr Cardiol Table 1 The age and pre-treatment hemodynamics during the headup test of both the effective and the ineffective groups Items
Effective group (n = 79)
Ineffective group (n = 25)
t, z or f value
p value
t = 0.32
0.75
t = -1.01
0.32
t = 0.81
0.42
Age (years)
11.8 ± 2.6
12.0 ± 2.4
DOS (mon)
6.5 ± 2.7
7.2 ± 3.7
SHR (bpm)
75.3 ± 10.3
73.3 ± 12.3
UHR (bpm)
108.9 ± 13.6
108.9 ± 13.6
t = -0.22
0.82
33.6 ± 11.6
34.9 ± 7.8
z = -0.03
0.98
Delta HR (bpm) SSBP (mmHg)
103.7 ± 9.3
102.4 ± 11.3
t = -0.54
0.59
USBP (mmHg)
104.0 ± 10.1
108.7 ± 12.4
t = 1.91
0.06
6.2 ± 10.7
z = -3.70
0.00*
Delta SBP (mmHg)
0.3 ± 6.9**
SDBP (mmHg)
63.6 ± 9.4
59.7 ± 10.0
t = -1.79
0.08
UDBP (mmHg)
67.4 ± 8.7*
72.6 ± 7.4
t = 2.70
0.01*
12.9 ± 7.0
z = -4.69
0.00*
5.0 ± 0.8
t = -0.53
0.595
4.7 ± 0.9
t = -15.6
0.00*
Delta DBP (mmHg)
3.8 ± 8.5**
SSBT
4.8 ± 0.9
SSAT
4
1.3 ± 0.9**
DOS duration of the symptoms, SHR supine heart rate, UHR upright heart rate, Delta HR the heart rate difference from the supine position to upright, SSBP supine systolic blood pressure, USBP upright systolic blood pressure, Delta SBP systolic blood pressure difference from the supine position to upright, SDBP supine diastolic blood pressure, UDBP upright diastolic blood pressure, Delta DBP diastolic blood pressure difference from the supine position to upright, SSBT symptom score before treatment, SSAT symptom score after treatment * p \ 0.05 versus ineffective group; ** p \ 0.01 versus ineffective group, p \ 0.01: compare symptom score between before treatment and after treatment in the effective group
To accurately predict the therapeutic efficacy of midodrine hydrochloride, an ROC curve analysis of the parallel association of SBP and DBP changes during the HUT, a model of combined analysis, was utilized. The results confirmed that midodrine hydrochloride was expected to be effective in children with POTS when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg, or when the increase in DBP from the supine position to standing during the HUT before treatment was B6.5 mmHg. Predictive sensitivity was 72 % and specificity was 88 %. In addition, ROC curve series connection of SBP and DBP changes, from the supine position to standing, revealed that the therapeutic efficacy of midodrine
123
Fig. 1 The ROC curve analysis based on the systolic pressure difference from the supine position to standing and the therapeutic efficacy of midodrine hydrochloride in children with POTS. Before treatment during the HUT, an ROC curve analysis was conducted in order to explore the suitability of using SBP changes from the supine position to standing to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.744 (95 % CI 0.616–0.871; p \ 0.05) and the cut-off value was 0 mmHg. The therapeutic efficacy of midodrine hydrochloride in children with POTS was hypothesized to be effective, when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg. The predictive sensitivity was 60 % and the predictive specificity was 80 %. The sensitivity is expressed on the longitudinal axis and 1-specificity on the horizontal axis. ROC receiver operating characteristic, POTS postural orthostatic tachycardia syndrome, HUT head-up test, SBP systolic blood pressure
hydrochloride in children with POTS was effective when the increase in SBP from the supine position to standing before treatment was B0 mmHg, and the increase in DBP from the supine position to standing before treatment was B6.5 mmHg. Predictive sensitivity was 58 % and specificity was 92 %. Comparison of the Effective Rate Based on the Cut-Off Values of Blood Pressure Changes During the Head-Up Test in Children with POTS. At the 6-month follow-up, the efficacy of midodrine hydrochloride in patients with an increase in SBP B 0 mmHg from the supine position to standing before treatment was significantly higher than those with an increase in SBP over 0 mmHg from the supine position to standing (90.3 vs. 61.5 %, p \ 0.01; Table 2). Similarly, the efficacy rate for midodrine hydrochloride in patients with an increase in DBP B 6.5 mmHg from the supine
Pediatr Cardiol Table 2 Comparison of the effective rate based on the cut-off values of SBP changes during the HUT in children with POTS Pre-treatment SBP increase from spine to upright
Effective cases (n)
Ineffective cases (n)
Total number (n)
Effective rate (%)
[0 mmHg
32
20
52
61.5
^0 mmHg
47
5
52
90.3*
SBP systolic blood pressure, HUT head-up test, POTS postural orthostatic tachycardia syndrome * p \ 0.01 versus [0 mmHg
Table 3 Comparison of the effective rate based on the cut-off values of DBP changes during the HUT in children with POTS Pre-treatment DBP increase from spine to upright
Fig. 2 The ROC curve analysis was based on the diastolic pressure difference from the supine position to standing and the therapeutic efficacy of midodrine hydrochloride in children with POTS. A ROC curve analysis was also conducted to explore the probability of using DBP changes from the supine position to standing before treatment during the HUT to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS. The area under the ROC curve was 0.809 (95 % CI 0.718–0.899; p \ 0.05) and the cut-off value was 6.5 mmHg. We confirmed that midodrine hydrochloride in children with POTS would be effective when the increase in DBP from the supine position to standing during HUT before treatment was B6.5 mmHg. The predictive sensitivity was 63 % and the predictive specificity was 92 %. The sensitivity is expressed on the longitudinal axis and 1-specificity on the horizontal axis. ROC receiver operating characteristic, POTS postural orthostatic tachycardia syndrome, DBP diastolic blood pressure, HUT head-up test
position to standing before treatment was significantly higher than those with an increase in DBP over 6.5 mmHg from the supine position to standing (95.0 vs. 64.1 %, p \ 0.01; Table 3).
Discussion In his study of nine patients treated with midodrine hydrochloride, Hoeldtke et al. [7] found that midodrine hydrochloride improved abnormal changes of the heart rate of patients with POTS during the HUT. Grubb et al. [5] believed that midodrine hydrochloride partially improved symptoms of OI. We previously compared the effective rate of different treatment modalities (midodrine hydrochloride plus oral rehydration salts, metoprolol plus oral rehydration salts, and oral rehydration salts only) and found that midodrine hydrochloride plus oral rehydration salts
Effective cases (n)
Ineffective cases (n)
Total number (n)
Effective rate (%)
[6.5 mmHg
41
23
64
64.1
^6.5 mmHg
38
2
40
95.0*
DBP diastolic blood pressure, HUT head-up test, POTS postural orthostatic tachycardia syndrome * p \ 0.01 versus [6.5 mmHg
were effective in some children with POTS [26]. In 2011, we studied 53 children with POTS treated with midodrine hydrochloride, and confirmed that there were some children with POTS who were unresponsive to midodrine hydrochloride [3]. Therefore, it is essential to be able to predict the efficacy of midodrine hydrochloride in children with POTS prior to initiating such therapy. The dose of midodrine hydrochloride used to treat POTS effectively is controversial, and there have been few studies indicating the standard dose of midodrine in pediatric medicine. In this study, the midodrine dose was 2.5 mg/day in accordance with previously published papers regarding pediatric cases [3, 9, 20, 25]. However, this dose seems relatively low and because of its short-acting nature, increasing the number of doses per day should be considered. In the future, additional studies are needed to clarify the use and dosage of midodrine in the pediatric population based on the fact that it is a short-acting drug. How to evaluate the therapeutic effect of midodrine hydrochloride on POTS was a difficult issue. There were several evaluating criteria, including the HUT outcome and symptoms, etc. As one of the criteria, the symptom score based on the typical symptoms of OI has been used; it is practical and simple and can quantify the therapeutic effect [27]. Therefore, in this study, we used this criterion to evaluate the therapeutic effect of midodrine hydrochloride in children with POTS, although using the symptom score together with symptom severity and repeated HUT would be more accurate.
123
Pediatr Cardiol
In this study, we discovered that the increase of standing SBP and DBP compared with the supine position before treatment in the effective group was markedly less than that of the ineffective group. Furthermore, ROC analysis showed that midodrine hydrochloride in children with POTS was expected to be effective when the increase in SBP from the supine position to standing during the HUT before treatment was B0 mmHg, or the increase in DBP from the supine position to standing during the HUT before treatment was B6.5 mmHg. Furthermore, the predictive sensitivity and specificity increased when using a combined modality parallel association ROC analysis of standing SBP change from the supine position, or standing DBP change from the supine position. However, predictive sensitivity for the series connection ROC analysis of standing SBP changes from the supine position to standing, and the DBP change from the supine position, was not favored, but specificity was high. Additionally, we found that the efficacy rate of midodrine hydrochloride in patients with increases in SBP or DBP below the cut-off values (from the supine position to standing, before treatment) was significantly higher than those with increases in SBP or DBP beyond the cut-off values. Based upon the above results, we reported that the therapeutic efficacy of midodrine hydrochloride in children with POTS could be predicted by measuring blood pressure during the HUT before treatment. Patients would get a better therapeutic response when the increase in SBP from the supine position to standing was B0 mmHg, or the increase in DBP from the supine position to standing was B6.5 mmHg. This information could greatly aid in the process of choosing which children with POTS would be responsive to midodrine hydrochloride. Recently, it was discovered that plasma MR-proADM of 61.5 % pg/ml produced both high sensitivity (100 %) and specificity (71.6 %) in predicting the efficacy of midodrine hydrochloride therapy for treating POTS in children [27]. Compared with plasma MR-proADM, measuring blood pressure had lower sensitivity (72 vs. 100 %) but higher specificity (88 vs. 71.6 %) in the prediction. At the same time, measuring blood pressure is simple, cheap, and noninvasive compared with the detection of MR-proADM levels. Therefore, using the SBP or DBP changes during the HUT before treatment to predict the therapeutic efficacy of midodrine hydrochloride in children with POTS is feasible and cost effective in clinical practice. The underlying mechanisms by which a relatively small increase or decrease or no change in blood pressure during the HUT predicts a better response to midodrine hydrochloride treatment in children with POTS, remain unclear. They might be associated with the pathogenesis of POTS and the consequent pharmacological effects of midodrine hydrochloride. The pathogenesis of POTS is complex and
123
presently unclear, but low blood volume, abnormal local blood flow, and autonomic stabilization of vascular tension are all hypothetical mechanisms [6, 13]. When children with POTS change position from the supine position to standing, their venous return is reduced, decreasing blood supply for the major organs, especially the brain, causing the OI symptoms to appear. The a1-adrenoreceptor agonist, midodrine hydrochloride, can increase the contractility of arterial vessels, maintaining the blood supply to the brain and thereby improving OI symptoms [17, 22]. Since the decreased venous return lowers the blood pressure, changes in blood pressure during the HUT seem useful in predicting the response to midodrine hydrochloride in children with POTS. The majority of cases with a relatively small increase or decrease or without any change in standing blood pressure versus supine blood pressure during the HUT test might be reflective of the suitability of the use of a1-adrenoreceptor agonists for children with POTS. However, although in each group some children responded well to the lowering of blood pressure during HUT, some did not. The mechanisms might involve the facts that midodrine hydrochloride as an a1-adrenoreceptor agonist functions via various pathways, including regulating the inappropriate control of peripheral vascular resistance that has been supposed to be a cause of POTS [12], decreasing sympathetic nervous activity and improving abnormal flow-mediated dilation in children with POTS [10]. In the study, the negative predictive values for SDP and DBP based on the predictive criterion were 38.5 and 36.0 %; namely, 61.5 and 64 % of patients who render the cut-off of the SDP and DBP, respectively, still respond to midodrine hydrochloride. The positive predictive values for SDP and DBP based on the predictive criterion were 90.3 and 95 %; namely, almost all patients who accord with the cut-off of the SDP and DBP, respectively, ideally respond to midodrine hydrochloride. Based on the above results, we suggested that children with POTS who accorded with the predictive criterion should, as the first choice, be treated with midodrine hydrochloride, while for those who render the predictive criterion, midodrine hydrochloride may be an alternative rather than the first choice. However, this present study has several limitations. Ideally, the inclusion of a placebo POTS group, where the patients’ outcome and prediction of BP change from the supine position to standing should have been analyzed and compared with the results of the present studies, would be meaningful for excluding the possibility of a natural history of symptoms with no relation to the midodrine. However, it was impractical to treat the children with POTS with placebo, without medication. Further studies including a placebo group are necessary. In this study, we only followedup the patients for 6 months, and were unable to know how
Pediatr Cardiol
long the therapeutic effect would last after the study period. Therefore, in the future, a long-term follow-up study of the value of hemodynamics in predicting the efficacy of midodrine hydrochloride in children with POTS is needed. In addition, a prospective cohort study for these criteria found in the present study should be designed in the future to see the generalizabilty of the findings. Acknowledgments This work was supported by the Major Basic Research Project of China (2012CB517806) and the National Twelfth Five-Year Plan for Science & Technology Support (2012BAI03B03). We thank Dr Jing Lin for her work on data processing. Conflict of interest
None.
References 1. Benrud-Larson LM, Dewar MS, Sandroni P, Rummans TA, Haythornthwaite JA, Low PA (2002) Quality of life in patients with postural tachycardia syndrome. Mayo Clin Proc 77:531–537 2. Chen L, Yang YY, Wang C, Wang HW, Tian H, Zhang QY, Chen JJ, Wang YL, Kang YL, Tang CS, Du JB (2006) A multicenter study of hemodynamic characteristics exhibited by children with unexplained syncope. Chin Med J 119:2062–2068 3. Chen L, Wang L, Sun J, Qin J, Tang C, Jin H, Du J (2011) Midodrine hydrochloride is effective in the treatment of children with postural orthostatic tachycardia syndrome. Circ J 75: 927–931 4. Chinese Pediatric Cardiology Society, Editorial Board of Chinese Journal of Pediatrics (2009) Guideline of diagnosis of syncope in children. Chin J Pediatr 47:99–100 5. Grubb BP, Karas B, Kosinski D, Boehm K (1999) Preliminary observations on the use of midodrine hydrochloride in the treatment of refractory neurocardiogenic syncope. J Interv Card Electrophysiol 3:139–143 6. Grubb BP, Kanjwal MY, Kosinski DJ (2001) The postural orthostatic tachycardia syndrome: current concepts in pathophysiology diagnosis and management. J Interv Card Electrophysiol 5(1):9–14 7. Hoeldtke RD, Bryner KD, Hoeldtke ME, Hobbs G (2006) Treatment of postural tachycardia syndrome: a comparison of octreotide and midodrine. Clin Auton Res 16:390–395 8. Julian MS, Amy W (2000) Vascular perturbations in the chronic orthostatic intolerance of the postural orthostatic tachycardia syndrome. J Appl Physiol 89:1505–1512 9. Kaufmann H, Saadia D, Voustianiouk A (2002) Midodrine in neurally mediated syncope: a double-blind randomized crossover study. Ann Neurol 52:342–345
10. Liao Y, Yang J, Zhang F et al (2013) Flow-mediated vasodilation as a predictor of therapeutic response to midodrine hydrochloride in children with postural orthostatic tachycardia syndrome. Am J Cardiol 112(6):816–820 11. Low PA, Sandroni P, Joyner M, Shen WK (2009) Postural tachycardia syndrome (POTS). J Cardiovasc Electrophysiol 20:352–358 12. Medow MS, Stewart JM (2007) The postural tachycardia syndrome. Cardiol Rev 15:67–75 13. Raj SR, Biaggioni I, Yamhure PC, Black BK, Paranjape SY, Byrne DW, Robertson D (2005) Renin-aldosterone paradox and perturbed blood volume regulation underlying postural tachycardia syndrome. Circulation 111:1574–1582 14. Rorbertson D (1999) The epidemic of orthostatic tachycardia and orthostatic intolerance. Am J Med Sci 317:75–77 15. Sandroni P, Tonette L, Benjamin R, Phillip AL (1999) Postural tachycardia syndrome: clinic features and follow-up study. Mayo Clin Proc 74:1106–1110 16. Schondorf R, Low PA (1993) Idiopathic postural orthostatic tachycardia syndrome: an attenuated form of acute pandysautonomia? Neurology 43(1):132–137 17. Sewarts JA (1988) Measuring the accuracy of diagnostic systems. Science 240:1285–1293 18. Steward JM (2002) Orthostatic intolerance in pediatrics. J Pediatr 140:404–450 19. Stewart JM (2004) Chronic orthostatic intolerance and the postural tachycardia syndrome (POTS). J Pediatr 145:725–730 20. Stewart JM (2006) Midodrine for the treatment of vasovagal syncope (simple faint). J Pediatr 149:740–742 21. Stewart JM, Gewitz MH, Weldon A, Munoz J (1999) Patterns of orthostatic intolerance: the orthostatic tachycardia syndrome and adolescent chronic fatigue. J Pediatr 135:218–225 22. Vaddadi G, Lambert E, Corcoran SJ, Esler MD (2007) Postural syncope: mechanisms and management. Med J Aust 187:299–304 23. Winder R, Barth A, Bidmon D, Ponocny I, Weber M, Mayr O, Robertson D, Diedrich A, Maier R, Pilger A, Haber P, Ru¨diger HW (2005) Endurance exercise training in orthostatic intolerance: a randomized controlled trial. Hypertension 45(3):391–398 24. Zhang QY, Du JB, Li WZ (2005) Clinical analysis and follow-up study of postural orthostatic tachycardia syndrome in 28 pediatric cases. Chin J Pediatr 43:165–169 25. Zhang Q, Du J, Tang C (2006) The efficacy of midodrine hydrochloride in the treatment of children with vasovagal syncope. J Pediatr 149:777–780 26. Zhang FW, Liao Y, Li XY, Chen L, Jin HF, Du JB (2011) An analysis of 118 cases for midodrine hydrochloride and metoprolol treating postural tachycardia syndrome in children. Chin J Pediatr 49:428–432 27. Zhang F, Li X, Ochs T, Chen L, Liao Y, Tang C, Jin H, Du J (2012) Midregional pro-adrenomedullin as predictor for therapeutic response to midodrinehydrochloride in children with postural orthostatic tachycardia syndrome. J Am Coll Cardiol 60:315–320
123
