ORIGINAL ARTICLE

Cardiac autonomic functions in children with familialMediterranean fever

Murat Şahin & Mustafa Kır & Balahan Makay &

Pembe Keskinoğlu & Elçin Bora & Erbil Ünsal &Nurettin Ünal

Received: 3 April 2014 /Revised: 13 May 2014 /Accepted: 3 June 2014# Clinical Rheumatology 2014

Abstract Familial Mediterranean fever (FMF) is the mostcommon inherited autoinflammatory disease in the world.The long-term effects of subclinical inflammation in FMFare not well recognized. Some studies have suggested thatFMF is associated with cardiac autonomic dysfunction inadult FMF patients. The objective of this study was to inves-tigate the cardiac autonomic functions in pediatric FMF pa-tients by using several autonomic tests. Thirty-five patientswith FMF and 35 healthy controls were enrolled in this cross-sectional study. Demographic data, disease-specific data, andorthostatic symptoms were recorded. In all participants, 12-lead electrocardiography (ECG), 24 h ambulatory electrocar-diographic monitoring, transthoracic echocardiography, tread-mill exercise test, and head upright tilt-table (HUTT) test wereperformed. The heart rate recovery (HRR) indices of the twogroups were similar. Also, chronotropic response was sim-ilar in both groups. The time-domain parameters ofheart rate variability (HRV) were similar in both groups,except mean RR (p=0.024). Frequencies of ventricularand supraventricular ectopic stimuli were similar in both

groups. There were no statistically significant differ-ences between the groups in average QT and averagecorrected QT interval length, average QT interval dis-persion, and average QT corrected dispersion. There wasno significant difference between the two groups regard-ing the ratio of clinical dysautonomic reactions onHUTT. However, we observed a significantly higher rateof dysautonomic reactions on HUTT in patients withexertional leg pain than that in patients without (p=0.013). When the fractal dimension of time curves werecompared, FMF patients exhibited significantly lowerdiastolic blood pressure parameters than controls in re-sponse to HUTT. Cardiovascular autonomic dysfunction inchildren with FMF is not prominent. Particularly, patients withexertional leg pain are more prone to have dysautonomicfeatures. Further studies are needed to elucidate the exactmechanisms leading to impaired cardiac autonomic functionsin FMF.

Keywords Autonomic Function . Familial Mediterraneanfever . Head upright tilt test

Introduction

Familial Mediterranean fever (FMF) is an autosomal recessivedisease characterized by recurrent attacks of fever and painfulinflammation of the peritoneum, synovium, or pleura. It is themost common inherited autoinflammatory disease worldwide[1]. Mutations in MEFV gene that encodes pyrin ormarenostrin are the genetic cause of the disease [2]. Themainstay of therapy is colchicine [3, 4]. Although currenttreatments lead to great improvement in the health status ofpatients with FMF, this disease may be associated with sig-nificant morbidity in some cases, such as amyloid nephropa-thy [5, 6]. Elevation of acute phase reactants during FMF

M. Şahin :M. Kır :N. ÜnalDepartment of Pediatrics, Division of Cardiology,Dokuz Eylül University Faculty of Medicine, Izmir, Turkey

P. KeskinoğluDepartment of Public Health, Dokuz Eylül University HealthSciences Institute, Izmir, Turkey

E. BoraDepartment of Medical Genetics, Dokuz Eylül University Hospital,Izmir, Turkey

B. Makay (*) : E. ÜnsalDepartment of Pediatrics, Division of Rheumatology,Dokuz Eylül University Hospital, 35340Balçova, Izmir, Turkeye-mail: balahan.bora@deu.edu.tr

Clin RheumatolDOI 10.1007/s10067-014-2714-z

attacks as well as attack-free periods is well documented in theliterature [7, 8]. The long-term effects of subclinical inflam-mation in FMF are not well recognized.

Autonomic nervous system (ANS) plays an important rolein the regulation of the cardiovascular system by providingoptimal function during several activities in healthy persons aswell as in mediating some manifestations of cardiac diseases.The cardiac ANS can be evaluated by several non-invasivemethods, such as heart rate recovery index (HRRI), heart ratevariability (HRV), heart rate turbulence (HRT), and QT dy-namics [9]. Previous studies reported a higher rate of abnor-mal cardiovascular reactivity and occult dysautonomia inadult FMF patients, on the basis of blood pressure, heart ratevalues, and cardiovascular reactivity score during a tilt test[10, 11]. On the other hand, Nussinovitch et al. found nodifference between FMF patients without amyloidosis contin-uously treated with colchicine and healthy controls regardingheart rate variability (HRV), which is a powerful and reliablemarker of ANS function [12]. There is no study evaluating thecardiac autonomic functions in pediatric FMF patients. Theobjective of this study was to investigate the cardiac autonom-ic functions in pediatric FMF patients using several autonomictests.

Study population

This cross-sectional study was conducted at Dokuz EylülUniversity Faculty of Medicine, Department of Pediatrics,Division of Rheumatology and Division of Cardiology, Izmir,Turkey, between January 2013 and March 2013. Participantsincluded children and adolescents (6 to 17 years old). Thirty-five patients (17 females and 18 males) with FMF and 35healthy controls (15 females and 20 males) were enrolled inthe study. The healthy controls were selected from the childrenwho visited the “well child outpatient clinic” of thedepartment.

The diagnosis of FMF was established according to theTel-Hashomer criteria [13]. All patients were on colchicine.Response to colchicine was evaluated according to previouslysuggested principles by Ben-Chetrit and Özdoğan dependingon the decrease rate of annual FMF attacks [14]. If colchicinereduced the attack rate by 50%, we designated it as “FMF-50”response. The patients who had an attack during the last2 weeks and who had concomitant chronic diseases such asjuvenile idiopathic arthritis were not included. Additionally,all the patients had normal urine analysis on follow-up visits,which likely excluded the diagnosis of amyloidosis.

Demographic data; age at diagnosis; disease duration; pres-ence of fever, abdominal pain, chest pain, joint pain, rash, andexertional leg pain (ELP); number of attacks in the last year;daily dose of colchicine; cumulative colchicine dose; diseaseseverity score; family history of FMF; MEFV mutation; and

serum C-reactive protein level were recorded for each patient.The disease severity score was calculated according to thescoring system suggested by Pras et al. [15].

Symptoms associated with dysautonomia such as repeatedfainting, orthostatic symptoms, decreased sweating, voidingdifficulty, chronic diarrhea, or constipation were recordedduring face-to-face interview with all participants. All partic-ipants were evaluated by transthoracic echocardiography inorder to exclude congenital heart defects prior to study.

The study protocol was approved by the ethics committeeof the Dokuz Eylul University, Faculty of Medicine in accor-dance with the Helsinki declaration. After the parents andchildren were informed about the study, written consents wereobtained.

Methods

Assessment of cardiovascular autonomic dysautonomia In allparticipants, we performed a 12-lead electrocardiography(ECG) at 25 mm/s (paper speed), 24-h ambulatory electrocar-diographic monitoring (by Lifecard CF Recorder, PathfinderECG analyzing Holter system), transthoracic echocardiogra-phy (by a Philips İE33 Medical System), treadmill exercisetest, and tilt-table test (by a Enraf-Nonius, Manumed tilt-tablesystem).

Heart rate recovery (HRR) indices

Heart rate recovery is defined as the rate of decrease in heartrate after a maximal graded exercise test. To calculate HRRindices, all subjects underwent exercise treadmill test in ac-cordance with the Bruce protocol [16]. After achieving peakworkload, all subjects spent at least 5 min of post-exerciseheart rate recorded without a cool-down period. Heart raterecovery indices were calculated by subtracting 1st, 2nd, 3rd,4th, and 5th minute heart rates from the maximal heart rateobtained during stress testing; these results were indicated asHRR1, HRR2, HRR3, HRR4, and HRR5, respectively.

Heart rate variability (HRV) parameters

By using a three-channel analog 24-h ambulatory electrocar-diographic monitoring, we assessed the heart rate variability(HRV) and the mean heart rate, QT corrected (QTc)/QT dis-persion analysis. In the HRVanalysis, the standard parametersobtained from the time-domain analysis of HRV include stan-dard deviation of all normal R-R intervals during 24 h(SDNN), standard deviation of all normal R-R intervals forall 5-min segments (SDNNi), standard deviation of the aver-age normal R-R intervals for all 5-min segments (SDANN),square root of the mean of the sum of the squares of

Clin Rheumatol

differences between adjacent R-R intervals (RMSSD), andintegral of the density distribution of N-N intervals dividedby the maximum of the density distribution (triangular index).

Tilt-table test

To compare the hemodynamic responses to autonomic chal-lenge, head upright tilt testing (HUTT) was performed in allsubjects. The protocol of the HUTTwas based on the 30-minsupine/30-min HUTT as previously described [17]. The mainoutcome measures were the values of systolic/diastolic/meanblood pressure (BP) and heart rate (HR) recorded duringrecumbence and tilt. The endpoints of vasodepressor andcardio-inhibitory reactions, orthostatic tachycardia, and pos-tural tachycardia syndrome were recorded.

Cardiovascular reactivity score (CVRS) for the assessment

A discriminant score referred to as CVRS was computedbased on BP and HR changes during HUTT. Systolic anddiastolic BP changes were defined as the difference betweenindividual BP values during HUTTand the last recumbent BPvalue, and divided by the last recumbent BP value. Heart ratechanges were defined as the difference between successiveHR values and the last recumbent HR value, and divided bythe last recumbent HR value. The BP and HR changes wererepresented graphically in time curves. These figures wereanalyzed by fractal dimension analysis on MATLAB andbox-count algorithm, as previously described [18]. The CVRSwas calculated in the group of FMF patients and healthycontrols. The best cutoff value was established, distinguishingthe cardiovascular reactivity of healthy subjects from that ofFMF patients. Sensitivity and specificity of the latter valuewere assessed.

Statistical analysis

Data were evaluated using the Statistical Package for SocialSciences (SPSS Inc, Chicago, USA) 16.0 program for Win-dows and by analyzing descriptive statistics (means, standarddeviation), comparing the means of quantitative data for dualgroups by using the Student t test and Mann-Whitney U testwhen appropriate. Chi-square test was used to evaluate thedifferences in proportions. Correlations between parameterswere computed through the Pearson’s correlation analysis.Correlation coefficient indicated low correlation at 0.10–0.29, moderate correlation at 0.30–0.49, and high correlationat ≥0.50. A p value ≤0.05 was considered as significant. TheCVRS was calculated by fractal dimension analysis. For thisanalysis, Excel and MATLAB (The Mathworks, Inc., 2012)programs were used.

Results

There were 18 boys and 17 girls with FMF and 20 boys and 15girls in the healthy control group (p=0.631). The mean age ofthe FMF patients was 11.6±3.5 years and control group was12.4±3.2 years (p=0.6). The mean BMI of the FMF patientswas 20±4.3 and control group was 20.2±3.8 (p=0.654). Nosignificant difference considering age, gender, and BMI wasfound between the patients and the controls. The most com-mon clinical manifestations of disease were abdominal pain(83 %) followed by fever (69 %). Besides, 47 % of patientscomplained of ELP. All the patients had FMF-50 response.The clinical characteristics of the patients were given inTable 1.

Data of genetic analysis were available in all patients. Themost prevalent genotype was M694V/M694Vamong patients(17 %). On face-to-face questioning, 54.3 % of FMF patientsreported orthostatic symptoms and two patients had history ofvasovagal syncope. The patients with ELP described signifi-cantly much more orthostatic symptoms than the patientswithout ELP (p<0.05). None of the healthy controls reportedorthostatic symptoms.

All the patients and controls had sinus rhythm and normal12-lead ECG results at rest. On transthoracic echocardiogra-phy, none of the participants had congenital heart disease.Comparison of cardiac autonomic function parameter valuesbetween FMF patients and healthy controls was summarizedin Table 2.

All study participants completed treadmill exercise stresstest without any complications. The HRR indices of the two

Table 1 Clinical characteristics of the FMF patients

Clinical characteristics Number Percentage

Variables

Fever 24 69

Abdominal pain 29 83

Chest pain 9 26

Rash 3 9

Arthritis/arthralgia 10 29

Family history of FMF 18 51

ELP 16 46

M694V homozygote 6 17

Mean SD

Disease duration (years) 5.6 2.8

Age at diagnosis (years) 8.5 4.0

Duration of colchicine use (years) 3.1 2.1

Dose of colchicine (mg/day) 0.96 0.29

Cumulative dose of colchicine (mg) 1,051 886

Severity score of FMF 6.2 1.5

CRP (mg/L) 2.4 4.3

ELP exertional leg pain, CRP C-reactive protein

Clin Rheumatol

groups were similar. Also, chronotropic responses were sim-ilar in both groups. We did not observe any chronotropicimpedance. Correlations between clinical characteristics ofFMF patients and cardiac autonomic function tests were sum-marized in Table 3. There was a remarkably negative correla-tion between chronotropic response and severity score of FMF(r=−0.396, p=0.018), and frequency of attacks (r=−0.426,p=0.012). Additionally, there was a moderate negative corre-lation between maximal heart rate and duration of colchicineuse (r=−0.402, p=0.017). Also, we showed that HRR1–5correlated negatively with duration of disease, duration ofcolchicine treatment, daily dose of colchicine, cumulativecolchicine dose, frequency of attacks, and FMF severity score.No correlations between HRR indices and initial presentingsymptoms of disease, genotype of disease, ELP, and C-reactive protein level were found (data not shown).

The time-domain parameters of HRV were studied in allparticipants. These results were compared between the FMFand control groups. On 24-h Holter ECG monitorization,HRV parameters were similar in the both groups, except meanRR (p=0.024; Table 2). There was a moderate correlationbetween frequency of attacks and RMSSD (r=−0.342, p=

0.048), and QTc (r=0.360, p=0.036). The other HRV param-eters showed no correlation with clinical characteristics of thedisease. Considering the rhythm abnormalities throughout theHolter ECG monitorization, we observed no severe arrhyth-mias including life-threatening ventricular ones in childrenwith FMF. Frequencies of ventricular and supraventricularectopic stimuli were similar in both groups (p=0.299). Therewere no statistically significant differences between thegroups in average QT and average corrected QT intervallength, average QT interval dispersion, and average QTcorrected dispersion.

There was no significant difference between the twogroups regarding the ratio of clinical dysautonomic reactionson HUTT. Eight FMF patients and five controls reachedclinical pathologic endpoints on HUTT (p=0.356). We ob-served a significantly higher rate of dysautonomic reactionson HUTT in patients with exertional leg pain (p=0.013).There were no correlations between dysautonomic reactionon HUTT and duration of disease, duration of colchicinetreatment, dose of colchicine, cumulative dose of colchicine,frequency of attacks, and FMF severity score (data notshown). Variables characterizing systolic/diastolic/mean BPand heart rate during the HUTT are summarized in Table 4.When the fractal dimension of time curves were compared,FMF patients exhibited significantly lower diastolic BP-sd(p=0.034) and lower diastolic BP-diff-a-fd (p=0.016) thancontrols in response to tilt. Multivariate analysis of all 21variables in the two groups identified the best predictors ofFMF patient vs healthy controls as DIAST-sd and DIAST-diff-a-fd. Based on the regression coefficients of these predic-tors, a linear discriminant score was computed for each sub-ject. This discriminant score was the CVRS.

CVRS ¼ 42:582−24:556� DIAST−diff−a−fdð Þ−0:125� ME−avgð Þ

Group averages with SD of CVRS were as follows: FMF0.334±0.762 and healthy controls −0.549±1.09 (p=0.001).We found that a CVRS of >0.05 was associated with FMF,with 74 % sensitivity and 77 % specificity.

Discussion

As no single test can provide global assessment of autonomicfunction, we have used several methods to investigate cardiacautonomic functions in children with FMF. Although previousreports showed several impaired autonomic functions in adultFMF patients, this study indicated that cardiovascular auto-nomic dysfunction in children with FMF was not prominent.To the best of our knowledge, this is the first study of cardiacdysautonomia in pediatric FMF patients using a comprehen-sive evaluation.

Table 2 Comparison of cardiac autonomic function parameters betweenFMF patients and healthy controls

FMF(mean±SD)

Control(mean±SD)

p value

Treadmill exercise test parameters

Basal heart rate (bpm) 105.8±13.2 105.1±15.6 0.850

Maximum heart rate (bpm) 183.5±8 182.6±11 0.694

Chronotropic response 77.6±13.9 77.4±13.8 0.943

HRR1 43.6±12.7 43.7±12.4 0.955

HRR2 57.8±11.4 58.8±13.2 0.744

HRR3 63.6±12.6 65.1±13.2 0.667

HRR4 65.4±19.6 67±12.4 0.653

HRR5 69.6±12.6 69.8±12.9 0.948

Heart rate variability parameters

Mean RR (msn) 699±67 740±81 0.024

SDNN (msn) 151±32 159±33 0.268

SDNNi (msn) 73±20 80±19 0.171

SDANN (msn) 130±31 135±31 0.543

RMSSD (msn) 60±28 65±24 0.447

Triangular index 42±12 43±10 0.732

QTc (msn) 421±16 414±14 0.760

QTd 29±9 27±8 0.438

HRR heart rate recovery, SDNN standard deviation of all normal R-Rintervals during 24 h, SDNNi standard deviation of all normal R-Rintervals for all 5-min segments, SDANN standard deviation of theaverage normal R-R intervals for all 5-min segments, RMSSD squareroot of the mean of the sum of the squares of differences between adjacentR-R intervals, QTc corrected QT, QTd dispersion QT

p<0.05 is significant

Clin Rheumatol

Heart rate recovery is particularly determined by cardio-vascular autonomic functions [19]. In a previous study, Ardıçet al. [20] reported that the HRR index was impaired in adultFMF patients with respect to healthy controls. Canpolat et al.[21] recently reported that adults with FMF showed delayedHRR compared to controls. They also showed that HRV,another powerful technique to evaluate autonomic functions,was abnormal in FMF patients [21]. In contrast ,Nussinovitch et al. [12] demonstrated that HRV parameterswere similar between FMF patients and healthy controls. Inthis study, we showed that both HRR and HRVof childrenwith FMF were similar to healthy peers. These controver-sial results between different studies may arise from thedifferent clinical, geographical, and genotypical character-istics of the patient populations. Besides, in this study, nostatistically significant differences were found between the

patients and healthy controls regarding QT dispersion andcorrected QT dispersion similar to many other studies in theliterature [22–25].

Adult studies in FMF patients exhibited an abnormal car-diovascular reactivity, which was clinically occult, but couldbe detected on autonomic challenge by HUTT [10, 11]. In thestudy of Rozenbaum et al., 18 % of patients experiencedclinical dysautonomic reactions during HUTT while none ofthe healthy controls had pathologic reactions [11]. We did notfind a significantly different rate of clinically pathologic reac-tions onHUTT between patients and controls. However, whenthe fractal dimension of time curves were compared, FMFpatients exhibited significantly lower diastolic BP parametersthan healthy controls in response to HUTT. Besides, thediscriminant score, CVRS, was significantly different be-tween the two groups. These results indicated that children

Table 3 Correlation analysis between clinical characteristics of FMF patients and cardiac autonomic function tests

Duration ofdisease

Duration offollow-up

Duration ofcolchicine use

Daily doseof colchicine

Cumulativecolchicine dose

Frequencyof attack

Severity scoreof FMF

Treamill exercise test parameters

Maximum heart rate −0.0430.806

−0.4020.017

−0.4080.015

−0.0390.825

−0.2060.235

−0.1550.381

0.3700.002

Chronotropic response −0.2130.219

−0.2710.115

−0.2520.145

−0.2960.084

−0.2810.102

−0.4260.012

−0.3960.018

HRR1 −0.3780.025

−0.2540.142

−0.2530.143

−0.5450.001

−0.3720.028

−0.3340.054

−0.3370.048

HRR2 −0.4580.006

−0.3810.024

−0.3770.025

0.597<0.001

−0.4820.003

−0.3180.067

−0.4940.003

HRR3 −0.3560.036

−0.3680.029

−0.3680.03

−0.5150.002

−0.4310.01

−0.3320.055

−0.5080.002

HRR4 −0.3530.037

−0.4380.008

−0.4350.009

−0.4690.005

−0.4890.003

−0.2480.157

−0.4980.002

HRR5 −0.3180.062

−0.3430.042

−0.3390.047

−0.583<0.001

−0.4350.009

−0.3430.047

−0.4750.004

Heart rate variability parameters

SDNN −0.0690.692

0.1840.290

0.1460.401

−0.0650,712

0.1410.421

−0.2430.167

0.1790.303

SDNNi −0.1580.364

0.1670.364

0.1240.477

−0.1150.509

0.0540.760

−0.2470.159

0.1200.491

SDANN −0.0570.746

−0.1180.498

0.0920.598

−0.1180.499

0.5900.094

−0.2370.176

0.0980.574

RMSSD −0.1680.334

0.1170.502

0.0800.648

−0.2470.153

−0.0100.956

−0.3420.048

0.0150.930

Triangular index −0.1130.520

0.1140.516

0.0720.679

0.2010.248

0.1530.380

−0.0720.685

0.1040.104

QTc −0.0270.887

−0.1580.364

−0.1640.347

−0.0890.610

−0.1980.254

0.3600.036

0.6740.070

QTd −0.0140.936

−0.0820.638

−.01030.555

0.1530.381

−0.0030.985

−0.0390.829

−0.0050.970

Above number indicate r value and below number indicate p value

HRR heart rate recovery, SDNN standard deviation of all normal R-R intervals during 24 h, SDNNi standard deviation of all normal R-R intervals for all5-min segments, SDANN standard deviation of the average normal R-R intervals for all 5-min segments, RMSSD square root of the mean of the sum ofthe squares of differences between adjacent R-R intervals, QTc QT corrected, QTd QT dispersion

p<0.05 is significant

Clin Rheumatol

with FMF might have occult dysautonomic cardiovascularreactivity.

The underlying mechanisms of the altered autonomic ner-vous activity in FMF patients were not clearly elucidated.Impaired autonomic functions were reported in several in-flammatory diseases such as rheumatoid arthritis, Sjögrensyndrome, systemic sclerosis, and systemic lupus erythema-tosus [26–29]. The asserted mechanisms for dysautonomia inthese diseases were neuropathy due to vasculitis, autonomicnerve dysfunction caused by inflammatory cytokines, disrup-tion of hypothalamo-pituitary-adrenal (HPA) axis, and ad-verse effects of disease-specific medications [26–30]. Indeed,the results of some studies considered the presence of adefective HPA axis in FMF patients [31, 32]. As metaraminol

infusion was shown to provoke an acute attack of FMF, it wassuggested that FMF is related to catecholamine metabolism[33]. Peripheral sympathetic afferents are directly activated bycirculating mediators and higher nervous centers that modu-late the cardiovascular reflexes. The other assumed mecha-nism, potential adverse effect of colchicine, is highly specu-lative, but some studies showed that the use of colchicine wasneurotoxic to cholinergic neurons in central and peripheralneuron systems [34, 35]. The negative correlation betweenheart rate recovery indices and duration and dose of colchicinein our study may lead a speculation to support the impact ofcolchicine on impaired autonomic functions.

Many FMF patients experience exercise-induced musclepain lasting for a few hours to the whole day, mainly affectingthe legs following physical exertion or even prolonged stand-ing, which is called ELP [13]. Another striking result of thisstudy was the significantly higher rate of clinicaldysautonomic reactions during HUTT in patients with ELP.This result raised the question whether mechanisms leading toELP in FMF patients might also cause impaired autonomicfunctions. Although increasingly recognized as a typical FMFmanifestation, the exact pathogenesis of ELP has not beenclearly understood yet. Dinç [36] suggested that increasedhydrostatic pressure in the lower extremities may be the mainfactor responsible for leg pain after prolonged standing andsitting. As leukocytes need adequate perfusion pressure topass through capillaries in microcirculation [37], the authorspeculated that catecholamines might increase the hydrostaticpressure of the capillary bed, which consequently lead to painat lower extremities. Previous studies showed that cardiacautonomic functions were impaired in patients with fibromy-algia as well [38, 39]. The most widely accepted mechanismfor this phenomenon was assumed as the high vascular sym-pathetic tone in patients with fibromyalgia as well as neuro-transmitter abnormalities [40].

The strength of this study is the inclusion of not only onebut also several powerful techniques to evaluate cardiovascu-lar autonomic system. Another strength is the assessment ofCVRS, which amplifies the contrast between the cardiovas-cular reactivity of patients and controls. The limitations of thestudy are its relatively small sample size and absence of apatient group not using colchicine.

In conclusion, this study suggested that cardiovascularautonomic dysfunction in children with FMF is not prominentas adult patients. Particularly, patients with exertional leg painare more prone to have dysautonomic features. Further studiesare needed to elucidate the exact mechanisms leading toimpaired cardiac autonomic functions in FMF.

Conflict of interest

None.

Table 4 Blood pressure and heart rate variables during head-up tilt test

Variables FMF (n=35) Control (n=35) p value

Mean SD Mean SD

SYST-avg 104.175 6.889 106.286 7.365 Ns

SYST-sd 5.527 1.802 5.740 1.769 Ns

SYST-diff-c-avg 0.005 0.030 0.019 0.037 Ns

SYST-diff-c-sd 0.053 0.017 0.055 0.018 Ns

SYST-diff-a-avg 0.042 0.017 0.049 0.021 Ns

SYST-diff-a-sd 0.040 0.017 0.044 0.016 Ns

SYST-diff-a-fd 1.267 0.024 1.276 0.023 Ns

DIAST-avg 66.233 6.895 67.014 5.714 Ns

DIAST-sd 7.246 2.726 8.755 2.518 0.034

DIAST-diff-c-avg 0.067 0.082 0.100 0.089 Ns

DIAST-diff-c-sd 0.121 0.056 0.147 0.054 Ns

DIAST-diff-a-avg 0.111 0.044 0.137 0.056 Ns

DIAST-diff-a-sd 0.107 0.052 0,130 0.057 Ns

DIAST-diff-a-fd 1.320 0.032 1.345 0.042 0.016

ME-avg 78.664 6.216 80.843 5.224 Ns

ME-sd 6.241 2.086 7.096 1.914 Ns

ME-diff-c-avg 0.059 0.076 0.060 0.046 Ns

ME-diff-c-sd 0.083 0.031 0.094 0.027 Ns

ME-diff-a-avg 0.085 0.064 0.083 0.031 Ns

ME-diff-a-sd 0.069 0.030 0.080 0.029 Ns

ME-diff-a-fd 1.300 0.026 1.307 0.025 Ns

HR-avg 90.169 8.545 91.507 10.905 Ns

HR-sd 12.287 5.283 13.223 5.629 Ns

HR-diff-c-avg 0.151 0.077 0.189 0.119 Ns

HR-diff-c-sd 0.161 0.076 0.182 0.100 Ns

HR-diff-a-avg 0.160 0.074 0.199 0.116 Ns

HR-diff-a-sd 0.152 0.076 0.171 0.101 Ns

HR-diff-a-fd 1.315 0.023 1.323 0.026 Ns

SYSTsystolic blood pressure,DIAST diastolic blood pressure,MEmedianblood pressure, avg average, diff differences, c current value, abs differ-ences expressed as absolute value, fd fractal dimension of BP differences,HR heart rate, Ns p>0.05, p<0.05 is significant

Clin Rheumatol

References

1. Samuels J, Ozen S (2006) Familial Mediterranean fever and the otherautoinflammatory syndromes: evaluation of the patient with recurrentfever. Curr Opin Rheumatol 18:108–117

2. Balow JE Jr, Shelton DA, Orsborn A et al (1997) A high-resolutiongenetic map of the familial Mediterranean fever candidate regionallows identification of haplotype-sharing among ethnic groups.Genomics 44:280–291

3. Zemer D, Pras M, Sohar E et al (1986) Colchicine in the preventionand treatment of the amyloidosis of familial Mediterranean fever. NEngl J Med 314:1001–1005

4. Zemer D, Revach M, Pras M et al (1974) A controlled trial ofcolchicine in preventing attacks of familial Mediterranean fever. NEngl J Med 291:932–934

5. Ludomirsky A, Passwell J, Boichis H (1981) Amyloidosis inchildren with familial Mediterranean fever. Arch Dis Child 56:464–467

6. Oner A, Erdoğan O, Demircin G et al (2003) Efficacy of colchicinetherapy in amyloid nephropathy of familial Mediterranean fever.Pediatr Nephrol 18:521–526

7. Bagci S, Toy B, Tuzun A et al (2004) Continuity of cytokineactivation in patients with familial Mediterranean fever. ClinRheumatol 23:333–337

8. Kiraz S, Ertenli I, Arici M, CalguneriM, Haznedaroglu I, Celik I, PayS, Kirazli S (1988) Effects of colchicine on inflammatory cytokinesand selectins in familial Mediterranean fever. Clin Exp Rheumatol16:721–724

9. TASK FORCE OF THE EUROPEAN SOCIETY OFCARDIOLOGY AND THE NORTH AMERICAN SOCIETY OFPACING AND ELECTROPHYSIOLOGY (1996) Heart rate vari-ability. Standards of measurement, physiological interpretation, andclinical use. Eur Heart J 17:354–381

10. RozenbaumM, Naschitz JE, Yudashkin M, Rosner I, Sabo E, ShavivN, Gaitini L, Zuckerman E, Yeshurun D (2002) Cardiovascularautonomic dysfunction in familial Mediterranean fever. JRheumatol 29:987–989

11. RozenbaumM,Naschitz JE, YudashkinM, Sabo E, Shaviv N, GaitiniL, Zuckerman E, Yeshurun D, Rosner I (2004) Cardiovascular reac-tivity score for the assessment of dysautonomia in familialMediterranean fever. Rheumatol Int 24:147–152

12. Nussinovitch N, Livneh A, Katz K, Langevitz P, Feld O,Nussinovitch M, Volovitz B, Lidar M, Nussinovitch U (2011)Heart rate variability in familial Mediterranean fever. RheumatolInt 31:39–43

13. Livneh A, Langevitz P, Zemer D, Zaks N, Kees S, Lidar T, Migdal A,Padeh S, Pras M (1997) Criteria for the diagnosis of familialMediterranean fever. Arthritis Rheum 40:1879–1885

14. Ben-Chetrit E, Ozdogan H (2008) Non-response to colchicine inFMF—definition, causes and suggested solutions. Clin ExpRheumatol 26(4 Suppl 50):49–51

15. Pras E, Livneh A, Balow JE Jr, Pras E, Kastner DL, Pras M,Langevitz P (1998) Clinical differences between North African andIraqi Jews with familial Mediterranean fever. Am J Med Genet 75:216–219

16. Bruce RA, Kusumi F, Hosmer D (1973) Maximal oxygen intake andnomographic assessment of functional aerobic impairment in cardio-vascular disease. Am Heart J 85:546–562

17. Naschitz JE, Sabo E, Naschitz S et al (2001) Hemodynamic instabil-ity in chronic fatigue syndrome: indices and diagnostic significance.Semin Arthritis Rheum 31:199–208

18. Naschitz JE, Sabo E, Naschitz S et al (2002) Fractal analysis andrecurrence quantification analysis of heart rate and pulse transit timefor diagnosing chronic fatigue syndrome. Clin Auton Res 12:262–274

19. Thomas DE, Exton SA, Yousef ZR (2010) Heart rate decelerationafter exercise predicts patients most likely to respond to cardiacresynchronisation therapy. Heart 96:1385–1389

20. Ardic I, Kaya MG, Yarlioglues M, Dogdu O, Celikbilek M, AkpekM, Akyol L, Torun E (2011) Rheumatol Int 31:121–125, Assessmentof heart rate recovery index in patients with familial Mediterraneanfever

21. Canpolat U, Dural M, Aytemir K, Akdoğan A, Kaya EB, Sahiner L,Yalçin U, Canpolat AG, CalgüneriM, Kabakçi G, Tokgözoğlu L, OtoA (2012) Evaluation of various cardiac autonomic indices in patientswith familial Mediterranean fever on colchicine treatment. AutonNeurosci 167:70–74

22. Koca B, Kasapçopur O, Bakari S, Sönmez E, Oztunç F, Eroğlu AG,Saltik L, Calay O (2012) QT dispersion and cardiac involvement inchildren with Familial Mediterranean fever. Cardiol Young 22(4):404–409

23. Topal F, Tanindi A, Kurtoglu HG, Akbulut S, Kucukazman M, TopalFE (2011) QT dispersion is not increased in familial Mediterraneanfever. J Int Med Res 39(5):2006–2011

24. Nussinovitch U, Kaminer K, Nussinovitch M, Volovitz B, Lidar M,Nussinovitch N, Livneh A (2012) QT interval variability in familialMediterranean fever: a study in colchicine-responsive and colchicine-resistant patients. Clin Rheumatol 31(5):795–799

25. Nussinovitch U, Livneh A, Volovitz B, Nussinovitch M, Ben-Zvi I,LidarM,Nussinovitch N (2012) Normal QT dispersion in colchicine-resistant familial Mediterranean fever (FMF). Clin Rheumatol 31(7):1093–1096

26. Janse van Rensburg DC, Ker JA, Grant CC, Fletcher L (2012)Autonomic impairment in rheumatoid arthritis. Int J Rheum Dis 15:419–426

27. Aydemir M, Yazisiz V, Basarici I, Avci AB, Erbasan F, Belgi A,Terzioglu E (2010) Cardiac autonomic profile in rheumatoid arthritisand systemic lupus erythematosus. Lupus 19:255–261

28. Newton JL, Frith J, Powell D, Hackett K,Wilton K, Bowman S, PriceE, Pease C, Andrews J, Emery P, Hunter J, Gupta M, Vadivelu S,Giles I, Isenberg D, Lanyon P, Jones A, Regan M, Cooper A, MootsR, Sutcliffe N, Bombardieri M, Pitzalis C, McLaren J, Young-Min S,Dasgupta B, Griffiths B, Lendrem D, Mitchell S, Ng WF, UKprimary Sjögren’s syndrome registry (2012) Autonomic symptomsare common and are associated with overall symptom burden anddisease activity in primary Sjogren’s syndrome. Ann Rheum Dis 71:1973–1979

29. Straub RH, Zeuner M, Lock G, Rath H, Hein D, Scholmerich J, LangB (1996) Autonomic and sensorimotor neuropathy in patients withsystemic lupus erythematosus and systemic sclerosis. J Rheumatol23:87–92

30. Demir H, Keleştimur F, Tunç M, Kirnap M, Ozügül Y (1999)Hypothalamo-pituitary-adrenal axis and growth hormone axis inpatients with rheumatoid arthritis. Scand J Rheumatol 28:41–46

31. Korkmaz C (2006) Is there a hypothalamic-pituitary-adrenal axisdysfunction in patients with familial Mediterranean fever? ClinRheumatol 25:606–607

32. Korkmaz C, Colak O, Alatas O, Ozarslan A, Ergul B (2002) Earlyblunted cortisol response to insulin induced hypoglycaemia in famil-ial Mediterranean fever. Clin Exp Rheumatol 20(4 Suppl 26):S8–S12

33. Huppertz HI, Michels H (1988) The metaraminol provocation test inthe diagnosis of familial Mediterranean fever. MonatsschrKinderheilkd 136:243–245

34. Peterson GM, McGinty JF (1988) Direct neurotoxic effects of col-chicine on cholinergic neurons in medial septum and striatum.Neurosci Lett 94:46–51

35. Ginn SR, Peterson GM (1992) Studies related to the use of colchicineas a neurotoxin in the septohippocampal cholinergic system. BrainRes 590:144–152

36. Dinç A (2000) Non-period leg pain in patients with familialMediterranean fever. Ann Rheum Dis 59:400

Clin Rheumatol

37. Adams DH, Nash GB (1996) Disturbance of leucocyte circulationand adhesion to the endothelium as factors in circulatory pathology.Br J Anaesth 77:17–31

38. Vallejo M, Martínez-Martínez LA, Grijalva-Quijada S, Olguín-Ruvalcaba HM, Salas E, Hermosillo AG, Cárdenas M, Martínez-Lavín M (2013) Prevalence of fibromyalgia in vasovagal syncope. JClin Rheumatol 19:111–114

39. da Cunha Ribeiro RP, Roschel H, Artioli GG, Dassouki T, PerandiniLA, Calich AL, de Sá Pinto AL, Lima FR, Bonfá E, Gualano B(2011) Cardiac autonomic impairment and chronotropic incompe-tence in fibromyalgia. Arthritis Res Ther 13(6):R190

40. Kulshreshtha P, Deepak KK (2013) Autonomic nervous systemprofile in fibromyalgia patients and its modulation by exercise: amini review. Clin Physiol Funct Imaging 33:83–91

Clin Rheumatol