Cardiovascular Toxicology

, Volume 13, Issue 3, pp 254–266 | Cite as

Impact of Opioid Pharmacotherapy on Arterial Stiffness and Vascular Ageing: Cross-Sectional and Longitudinal Studies

Article

Abstract

Whilst there is a small literature on the cardiovascular toxicity of opiates, there is no detailed antemortem data on non-cardiovascular patient populations. A cross-sectional and longitudinal naturalistic observational study was performed comparing methadone (N = 71)-, buprenorphine (N = 593)-, naltrexone (N = 23)-treated patients with controls (N = 576) on indices of arterial stiffness and vascular age by Pulse Wave Analysis in primary care, 2006–2011. Controls were younger 29.96 ± 0.45 (mean ± SEM) vs. 34.00 ± 0.34–39.22 ± 1.11 years (all P < 0.005) and had fewer smokers (15.9 % vs. 86.9 %–92.96 %, all P < 0.0001). The sex ratio was similar (69.6 vs. 67.7 % male, P = 0.46). These baseline differences were controlled for by multiple regression. Linear regression of vascular age, central augmentation pressure, central augmentation index and other measures against chronologic age showed significant protective effects by treatment group against the treatment standard of methadone, in both sexes in additive and interactive models (all P < 0.02). Interactive terms in treatment type remained significant including all conventional risk factors accounting for differing opiate exposures. The principal findings from multiple regression were confirmed in the time series analysis up to 5 years by repeated measures nonlinear regression. These studies show that the deleterious impact of chronic opiate pharmacotherapy on vascular age and arterial stiffness varies significantly by treatment type.

Keywords

Arterial stiffness Heroin Opiates Vascular ageing Methadone Buprenorphine Naltrexone 

Introduction

Opiate dependence is an increasingly common disorder which may arise via multiple pathways including chronic pain management, iatrogenically or from overt drug addiction to either licit or illicit narcotics. In 2007, in USA, almost 100 persons died of prescription drug overdose daily, and in 2008, there were 14,800 deaths due to opiates [1]. This was more than was associated with illicit heroin and cocaine lethal toxicity combined. Other data show a lifetime report of 7.3 % of drug dependence amongst surveyed Americans, with 235 million opiate scripts written in 2004 in USA [2]. A total of 90 million Americans complain of chronic pain, a condition associated with annual healthcare costs of over $100 billion. Opiate dependence is typically long-term over several years and refractory [3, 4].

In addition to acting via well-described classical (μ, δ and κ) and non-classical (nociceptin/orphanin and sigma) receptors which signal via a gyanyl cyclase system coupled to CREB (cAMP response element binding protein), opiates act via multiple other pathways [5]. Of great interest to cardiovascular physicians is literature dating back over 40 years showing that opiates directly affect cellular and tissue growth [6] principally by an action mediated by a perinuclear receptor for Met-enkephalin known as the opiate growth factor receptor which signals via P16INK4A and P21WAF1/CIP1 [7]. P16 is transcribed from the gene CDKN2A near the chromosome 9p21.3 region identified on many genome-wide association studies as being linked with cardiovascular diseases, diabetes and other degenerative disorders [8, 9, 10, 11]. Activity at this site has since been shown to be linked with effects mediated by a long non-coding RNA known as ANRIL (anti-sense non-coding RNA in the senescence locus) [12, 13]. Opiate dependence and its management by long-term maintenance on opiate agonist (methadone), partial agonist (buprenorphine) or antagonist (naltrexone) therefore represent an ideal opportunity to consider and study the role of tissue trophic agents and agents acting via the cell cycle and the impact of the senescence locus on chromosome 9p21 in cardiovascular disease.

There is a small but intriguing literature related to the stimulatory effect of chronic opiate therapy on atherosclerotic coronary artery disease. One large data linkage study of Australia’s most populous state (New South Wales) involving over 42,000 patients and 431,000 patient-years of observation and 21-year experience found an elevated risk of coronary disease in opiate dependence of 2.2-fold (95 % CI 1.8–2.7) [3]. Similar results were also published from Iran where a large study recently found an elevation of the adjusted hazard ratio of death from coronary artery disease of 1.90 (1.57–2.29) and of cerebrovascular disease of 1.68 (1.29–2.28), both much worse in females [14]. The Sydney group identified that 17.3 % of narcotic decedents had a coronary stenosis of more than 75 % over the age of 44 years [15]. They also found a higher rate of major system disease and multiple systemic diseases [15]. A similar finding was made in a clinical angiographic study where an adjusted odds ratio of 1.8 (CI 1.1–3.1) and a dose–response effect were observed with the severity of coronary artery disease [16]. These workers have also noted that opium addicts have coronary surgery 4 years earlier than non-dependent patients [17]. Elevated rates of atherosclerotic risk factors have also been noted in opiate dependence including poor diet, hyperglycaemia and diabetes [18], hypertension [19], dyslipidaemia [20] and weight gain [21].

The New South Wales group has further shown that methadone exacerbates cardiac (adjusted OR = 3.13, CI 2.00–4.90) and all system (OR = 3.43, CI 1.65–7.14) pathology compared to untreated heroin addiction [22]. Treatment with long-term agonists particularly methadone has been shown to prolong opiate dependency syndromes by up to five times [4]. Differential effects have also been noted in methadone-treated patients in relation to their dental health [23]. Chronic dental infection has been linked with atherosclerotic disease, and the antigens of oral pathogens and the microorganisms themselves have been cultured from atherosclerotic plaque and coronary arteries [24]. It is conceptually plausible that the chronic immune stimulation involved with chronic periodontitis, bone loss from the jaws and its associated severe infection is associated with various adverse health outcomes.

Contrariwise opiate antagonists such as naltrexone have been associated with a reversal of the negative tissue trophic activities of opiate agonists [25]. Buprenorphine is a partial opiate agonist and may lie intermediate on the spectrum between full opiate agonists and antagonists. Since the use of these three agents has been described clinically for the treatment of opiate dependence, the real possibility exists that significant differences will be associated with their use clinically in patients and particularly in the disease spectrum encountered after long-term treatment.

Pulse wave analysis (PWA) by radial arterial applanation tonometry is a technique which has been used to study central vascular health. It allows the calculation of parameters of central vascular stiffness and central blood pressures against which the heart must work from a deconvolution and back extrapolation from averaged high-fidelity pressure waveforms taken from the radial artery at the wrist standardized against brachial blood pressure [26, 27]. Central blood pressure derived from the forward wave from the cardiac cycle is augmented by a backwards reflected waveform emanating from resistance sites in the peripheral circulation. This pressure augmentation has been related to vascular age by population-based studies. The central augmented pressure is related linearly to vascular age and, when indexed to the pulse height, is related in a curvilinear, convex upwards, manner to vascular age. The vascular age is calculated as part of the algorithm included with the SphygmoCor software from Atcor Medical supplies [28].

As this clinic sees both opiate-dependent and non-dependent patients and is experienced in the use of the pulse wave analysis technique, we are in an ideal situation to compare the effects of treatment with these various pharmacotherapies both cross-sectionally and longitudinally on validated cardiovascular subclinical endophenotypes in living patients. The hypothesis was generated before the study was launched.

Methods

Patient Selection. Healthy control patients (N = 576: 69.62 % male) were studied opportunistically when they presented for health check-ups for employment or sporting medicals, or for minor conditions which were not known to perturb the cardiovasculature such as ear wax removal or minor psychological complaints. University students were also sampled. Patients with known chronic cardiovascular conditions such as diabetes or hypertension were excluded, as were those with conditions known to perturb the cardiovasculature acutely or sub-acutely such as infectious disorders. The study period was 2006–2011.

Treatment. Opiate-dependent patients presenting who were stabilized on methadone (N = 71: 57.75 % male) or buprenorphine (N = 593: 68.80 % male) were studied opportunistically if they were physically otherwise well. These are both registered treatments for opiate dependence in Australia. Patients were treated by their usual treating physician according to established clinical practices. In Australia, naltrexone implants are available on a compassionate basis for patients with life-threatening conditions through the Australian Regulatory Authority (Therapeutic Goods Administration) Special Access Scheme. These were used via a technique previously described [29]. In brief, naltrexone implants were sourced from Perth’s “Go Medical Industries”. They were inserted surgically via a small skin incision usually in the left iliac fossa under local anaesthetic after patients had been abstinent for at least 48 h with patients receiving symptomatic withdrawal medication as required. A total of 23 patients (69.57 % male) recovered quickly within a few hours and were discharged.

PWA Testing. PWA testing was performed usually on the right radial artery by the applanation tonometry technique, using the “Miller” microtonometer, the SphygmoCor software and preamplifier and the Atcor Medical system from Sydney. Patients were rested for 5 min prior to study and were not allowed to talk while the study was in progress. Eating, drinking and alcohol consumption were not restricted. However, if patients were known to have been drinking prior to study, their study was discounted. The brachial blood pressure was sampled in the arm opposite to that in which the study was performed using an Omron HEM 907 oscillometric device. Notes were taken at the time of study and entered into the software of tobacco, drug and opiate use. All studies were performed in quintuplicate and averaged on each day. Studies which were found to be either “inconclusive” by the software analytical system or had an operator index less than 70 were disregarded.

Major indices calculated from this technique included the Vascular or Reference Age (VA, RA), the Chronologic Age (CA), the Central Augmentation Pressure at Heart Rate 75 (C_AP_HR75), the Central Augmentation Pressure/Pulse Height Ratio at Heart Rate 75 (C_AGPH_HR75) also known as the Augmentation Index, Central Pulse Height (C_PH), Peripheral–Central Pulse Pressure Amplification Ratio (PPAmpRatio), Central Systolic Pressure (C_SP), Central Diastolic Pressure (C_DP), Central Mean Pressure (C_MEANP), Central End Systolic Pressure (C_ESP), the Central Diastolic Time Index (C_DTI), the Central Tension Time index (C_TTI), the Central Diastolic Duration (C_DD), and an index of subendocardial perfusion known variously as the Subendocardial Perfusion Ratio (SEVR), the Central Stroke Volume Index (C_SVI) or the Buckberg ratio, which is defined as the C_TTI/C_DTI. The physiology and relationship of these various stiffness indices to chronological age is explained further in the “Discussion” section.

Statistics. Data are presented as mean ±SEM. Categorical variables were compared using the chi-square test in EpiInfo 7.0.8.3 from Atlanta, Georgia. If there were less than 20 in a cell, Fisher’s exact test, also in EpiInfo, was used. Continuous variables were compared using “Statistica” 7.1 from Statsoft, Oklahoma. Variables were log or arcsinh transformed in the interests of normality. Arcsinh is a transformation very close to a logarithmic transformation, which also accepts zero or negative arguments. It was employed in models which included CRP. Models were chosen based on the minimization of the Akaike Information Criterion (AIC). Multiple regression and repeated measures nonlinear mixed effects regression were conducted in “R” 2.13.1 obtained from the Central “R” Archive Network mirror at the University of Melbourne. In repeated measures regressions, time and the patient identification code were treated as random effects. Missing data points were omitted from all analyses. Final models are presented in each case after model reduction by the classical method of omitting the least significant terms. Graphs were drawn using ggplot2 software in “R”. P < 0.05 was regarded as significant.

Ethical Approval. All patients were consented prior to their inclusion in the study. Patients who had a naltrexone implant inserted consented in writing. Regulatory compliance with relevant state and Federal legislation was maintained for all patients. Patient confidentiality was respected throughout. The study was approved by the Human Research Ethics Committee (HREC) of the Southcity Medical Centre which is a nationally approved HREC from the National Health and Medical Research Council. All procedures were in accord with the Declaration of Helsinki.

Results

It may be of assistance to the reader to set out plainly the plan for the presentation of the results. Baseline and socio-demographic data are presented in the text and Table 1. Between-group comparisons for key cardiovascular variables are shown in Supplementary Table 1, whilst similar data are presented for age-, height-, weight-, BMI- and opiate exposure-matched groups in Supplementary Table 2. Supplementary Table 3 presents similar data for RA adjusted for basic clinical variables. Cross-sectional data for age are presented graphically in Fig. 1, and for arterial stiffness, pressure and timing in Supplementary Figures 1–3. These data are quantitated and summarized in Tables 2 and 3 for additive and interactive models, respectively. Multiple regression of log (RA/CA) is introduced in Supplementary Table 4. Table 4 presents a formal multiple regression of all CVS variables against log (RA/CA). Figure 2 presents the longitudinal data which is quantitated in Table 5. Table 6 presents data from a more extensive repeated measure multiple regression analysis, thereby adding the depth of longitudinal data to the detail of cross-sectional data.
Table 1

Socio-demographic data

Parameters

Data

P values

Control

Buprenorphine

Methadone

Naltrexone

Ctl-Bup

Ctl-Mdn

Ctl-Ntx

Bup-Mdn

Bup-Ntx

Mdn-Ntx

No. of patients

576

593

71

23

      

No. of studies

710

1147

83

80

      

Timing of repeat PWA

680.83 (41.22)

574.68 (21.78)

787.87 (129.79)

294.02 (49.80)

      

Male sex—no. (%)

401 (69.62 %)

408 (68.8 %)

41 (57.75 %)

16 (69.57 %)

0.7628

0.0427

1.0000

0.0601

1.0000

0.3395

Biometrics

Chronologic_agea

29.96 (0.45)

34.00 (0.34)

39.22 (1.11)

34.22 (1.82)

0.0000

0.0000

0.0035

0.0000

0.8748

0.0248

Height (cm)

173.59 (0.37)

173.62 (0.35)

171.7 (1.13)

176.17 (1.26)

0.9580

0.0935

0.0593

0.0811

0.0612

0.0104

Weight (kg)

73.21 (0.58)

74.55 (0.67)

77.59 (2.16)

78 (3.41)

0.1296

0.0537

0.1066

0.1434

0.3204

0.9240

BMI (kg/m2)

24.25 (0.17)

24.62 (0.18)

26.24 (0.63)

25.1 (1.02)

0.1414

0.0033

0.3286

0.0161

0.6122

0.3664

Drug use

Smokers

15.97 %

92.41 %

92.96 %

86.96 %

0.0000

0.0000

0.0000

0.8693

0.3391

0.3755

Cigarettes/day

1.78 (0.22)

16.93 (0.39)

15.35 (1.26)

13.26 (1.71)

0.0000

0.0000

0.0000

0.1913

0.0678

0.3885

Min. post-cigarette

121.98 (8.48)

73.5 (9.43)

144.35 (37.78)

420.47 (228.06)

0.0001

0.5652

0.2093

0.0732

0.1479

0.2488

Heroin use (Ever)

0.01 (0)

0.99 (0)

0.96 (0.02)

0.87 (0.07)

0.0000

0.0000

0.0000

0.1688

0.1038

0.2543

Opiate dose (g/day)

0 (0)

0.49 (0.02)

0.94 (0.17)

0.39 (0.07)

0.0000

0.0000

0.0000

0.0112

0.3225

0.0040

Opiate Dura’n (years)

0.09 (0.05)

12.53 (0.33)

18.29 (1.36)

9.9 (1.65)

0.0000

0.0000

0.0000

0.0001

0.1398

0.0003

Opiate dose duration (gayears)

0.03 (0.02)

6.3 (0.3)

22.19 (5.71)

4.43 (1.35)

0.0000

0.0002

0.0039

0.0071

0.2487

0.0034

Laboratory data

Cholesterol (mmol/l)

4.78 (0.08)

4.41 (0.04)

5.16 (0.25)

4.39 (0.24)

0.0001

0.0817

0.1274

0.0054

0.9423

0.0395

Triglyceride (mmol/l)

1.43 (0.08)

1.37 (0.04)

1.9 (0.32)

1.33 (0.22)

0.3977

0.0397

0.6594

0.1097

0.8487

0.1989

HDL (mmol/l)

1.31 (0.04)

1.23 (0.02)

1.27 (0.06)

1.43 (0.15)

0.0742

0.7090

0.2744

0.5973

0.0349

0.2852

LDL (mmol/l)

2.87 (0.1)

2.47 (0.04)

2.59 (0.25)

2.353125 (0.19)

0.0003

0.2481

0.0440

0.6446

0.5824

0.4799

ALT (IU/l)

32.8 (2.07)

66.73 (4.4)

46.71 (4.87)

66.65 (15.99)

0.0000

0.0116

0.0471

0.2568

0.9972

0.2434

Creatinine (mmol/l)

78.86 (1.28)

79.88 (0.7)

80.5 (3.18)

75.48 (3.67)

0.4689

0.6185

0.3841

0.8321

0.2086

0.3109

CRP (mg/l)

3.43 (0.65)

6.23 (0.46)

6.01875 (1.36)

2.8234375 (1.07)

0.0005

0.1217

0.7583

0.9123

0.1523

0.0994

Ctl control, Bup buprenorphine, Mdn methadone, Ntx naltrexone

aStatistics for log transformed data presented

Fig. 1

Ageing indices by chronologic age by treatment type

Table 2

Age-dependent multiple regression of key cardiovascular variables—additive models

Parameters

Variables

Parameter estimates

t value

Pr(>|t|)

Estimate

SE

All patients

Vascular age

Control

−0.2651

0.042

−6.282

4.59E−10

Vascular age

Buprenorphine

−0.2374

0.041

−5.724

1.30E−08

Vascular age

Naltrexone

−0.2368

0.079

−3.004

0.0027

C_AP_HR75

Buprenorphine

−3.2025

0.490

−6.532

9.4E−11

C_AP_HR75

Control

−3.1518

0.502

−6.285

4.5E−10

C_AP_HR75

Naltrexone

−2.9867

0.932

−3.204

0.0014

C_AGPH_HR75

Buprenorphine

−6.338

1.262

−5.023

5.8E−07

C_AGPH_HR75

Control

−6.524

1.291

−5.055

4.9E−07

C_AGPH_HR75

Naltrexone

−5.914

2.399

−2.466

0.0138

C_SP

Buprenorphine

−6.825

1.320

−5.169

2.7E−07

C_SP

Control

−5.216

1.350

−3.862

0.0001

C_SP

Naltrexone

−5.245

2.510

−2.09

0.0368

C_SVI

Buprenorphine

7.431

3.777

1.967

0.0494

C_SVI

Control

9.644

3.863

2.496

0.0127

Males

Vascular age

Buprenorphine

−0.2814

0.0537

−5.2390

0.0000

Vascular age

Control

−0.2935

0.0550

−5.3400

0.0000

Vascular age

Naltrexone

−0.2877

0.0965

−2.9810

0.0030

C_AP_HR75

Buprenorphine

−3.4480

0.5948

−5.7970

0.0000

C_AP_HR75

Control

−3.1768

0.6120

−5.1900

0.0000

C_AP_HR75

Naltrexone

−3.2355

1.0685

−3.0280

0.0025

C_AGPH_HR75

Buprenorphine

−7.0490

1.4940

−4.7170

0.0000

C_AGPH_HR75

Control

−6.7190

1.5380

−4.3690

0.0000

C_AGPH_HR75

Naltrexone

−6.2850

2.6850

−2.3410

0.0195

C_SP

Buprenorphine

−6.8630

1.7030

−4.0300

0.0001

C_SP

Control

−5.1450

1.7520

−2.9360

0.0034

C_SVI

Buprenorphine

14.2480

5.0720

2.8090

0.0051

C_SVI

Control

17.0380

5.2180

3.2650

0.0011

Females

Vascular age

Buprenorphine

−0.1595

0.0661

−2.4110

0.0164

Vascular age

Control

−0.2335

0.0664

−3.5180

0.0005

C_AP_HR75

Buprenorphine

−1.8716

0.7273

−2.5740

0.0104

C_AP_HR75

Control

−2.5167

0.7337

−3.4300

0.0007

C_AGPH_HR75

Control

−4.4860

1.8900

−2.3740

0.0181

C_SP

Buprenorphine

−7.0600

2.0730

−3.4060

0.0007

C_SP

Control

−6.2470

2.0910

−2.9870

0.0030

Abbreviations as in “Methods” section

Table 3

Age-dependent multiple regression of key cardiovascular variables—interactive models

Parameters

Variables

Parameter estimates

t value

Pr(>|t|)

Estimate (SE)

All patients

Vascular age

CA:Control

−0.0067 (0.0011)

−6.2250

0.0000

Vascular age

CA:Buprenorphine

−0.0058 (0.001)

−5.5240

0.0000

Vascular age

CA:Naltrexone

−0.0055 (0.0022)

−2.5480

0.0110

C_AP_HR75

CA:Buprenorphine

−0.8993 (0.1349)

−6.6680

0.0000

C_AP_HR75

CA:Control

−0.8745 (0.1384)

−6.3170

0.0000

C_AP_HR75

CA:Naltrexone

−0.8122 (0.2627)

−3.0910

0.0020

C_AGPH_HR75

CA:Buprenorphine

−1.7589 (0.3473)

−5.0650

0.0000

C_AGPH_HR75

CA:Control

−1.7863 (0.3564)

−5.0110

0.0000

C_AGPH_HR75

CA:Naltrexone

−1.5927 (0.6766)

−2.3540

0.0187

C_SP

CA:Buprenorphine

−15.839 (5.305)

−2.9860

0.0029

C_SVI

CA:Buprenorphine

2.0630 (1.0400)

1.9840

0.0475

C_SVI

CA:Control

2.6370 (1.0670)

2.4710

0.0136

Males

C_AP_HR75

CA:Buprenorphine

−0.9573 (0.1635)

−5.8560

0.0000

C_AP_HR75

CA:Control

−0.8774 (0.1691)

−5.1880

0.0000

C_AP_HR75

CA:Naltrexone

−0.8946 (0.3039)

−2.9440

0.0033

C_AGPH_HR75

CA:Buprenorphine

−7.049 (1.494)

−4.7170

0.0000

C_AGPH_HR75

CA:Control

−6.719 (1.538)

−4.3690

0.0000

C_AGPH_HR75

CA:Naltrexone

−6.285 (2.685)

−2.3410

0.0195

C_SP

CA:Buprenorphine

−1.9547 (0.4678)

−4.1790

0.0000

C_SP

CA:Control

−1.4092 (0.4839)

−2.9120

0.0037

C_SVI

CA:Buprenorphine

3.8993 (1.3942)

2.7970

0.0053

C_SVI

CA:Control

4.7618 (1.4424)

3.3010

0.0010

Females

Vascular age

CA:Buprenorphine

−0.0037 (0.0017)

−2.2180

0.0271

Vascular age

CA:Control

−0.0061 (0.0017)

−3.6800

0.0003

C_AP_HR75

CA:Buprenorphine

−0.5311 (0.2007)

−2.6470

0.0085

C_AP_HR75

CA:Control

−0.7182 (0.2021)

−3.5540

0.0004

C_AGPH_HR75

CA:Control

−1.2775 (0.521)

−2.4520

0.0146

C_SP

CA:Buprenorphine

−2.0233 (0.5719)

−3.5380

0.0005

C_SP

CA:Control

−1.7589 (0.5759)

−3.0540

0.0024

Abbreviations as in “Methods” section

Table 4

Multiple regression of all CVS risk factors and pharmacotherapy for log (RA/CA)

Parameters

Estimate (SE)

Pr(>|t|)

All patients

SP

8.2190 (2.08)

0.0001

SP:CRP:MaleSex:RxNaltrexone

−0.1864 (0.05)

0.0001

SP:RxBuprenorphine

−7.7511 (2.11)

0.0003

RxBuprenorphine

37.6763 (10.25)

0.0003

RxControl

37.8602 (10.61)

0.0004

SP:RxControl

−7.7822 (2.18)

0.0004

Height

−0.0073 (0)

0.0007

CRP

19.2222 (5.79)

0.0010

SP:CRP

−3.9762 (1.2)

0.0010

CRP:RxBuprenorphine

−18.8199 (5.83)

0.0013

SP:CRP:RxBuprenorphine

3.8934 (1.21)

0.0013

Cigarettes:HDL:CRP:MaleSex

−0.1701 (0.05)

0.0017

CRP:RxControl

−19.3973 (6.16)

0.0017

SP:CRP:RxControl

4.0041 (1.28)

0.0018

SP:Cigarettes:HDL:CRP:MaleSex

0.0341 (0.01)

0.0020

SP:Cigarettes:HDL:CRP:FemaleSex

0.0954 (0.03)

0.0026

Cigarettes:HDL:CRP:FemaleSex

−0.4521 (0.15)

0.0026

SP:CRP:MaleSex

0.043 (0.02)

0.0199

SP:CRP:MaleSex:RxBuprenorphine

−0.0405 (0.02)

0.0299

SP:CRP:MaleSex:RxControl

−0.0427 (0.02)

0.0380

SP:RxNaltrexone

−5.4745 (2.9)

0.0595

RxNaltrexone

26.3429 (14.07)

0.0618

Males

Cigs:HDL:CRP:Buprenorphine

−0.0282 (0.0113)

0.0130

SP:Cigs:HDL:CRP:Buprenorphine

0.0002 (0.0001)

0.0178

Females

Cigs:HDL

−0.2143 (0.0907)

0.0196

Cigs:HDL:Buprenorphine

0.2043 (0.0909)

0.0262

Cigs:HDL:Naltrexone

0.1904 (0.0924)

0.0413

SP systolic pressure

Fig. 2

Ageing indices by arcsinh time by treatment type

Table 5

Repeated measures regression of pharmacotherapy and age for log (RA)

Variables

Estimate (SE)

df

t value

P value

AIC

Loglik

All patients

Buprenorphine

−0.2462 (0.0416)

627

−5.9191

0.0000

834.46

−409.23

Naltrexone

−0.3122 (0.0592)

627

−5.2757

0.0000

834.46

−409.23

CA:Buprenorphine

−0.0674 (0.0114)

627

−5.9206

0.0000

839.50

−411.75

CA:Naltrexone

−0.0858 (0.0165)

627

−5.2160

0.0000

839.50

−411.75

Males

Buprenorphine

−0.2682 (0.0535)

434

−5.0185

0.0000

561.81

−272.91

Naltrexone

−0.3304 (0.0719)

434

−4.5971

0.0000

561.81

−272.91

Females

Buprenorphine

−0.1819 (0.0673)

177

−2.7029

0.0075

289.87

−136.93

Naltrexone

−0.2658 (0.1049)

177

−2.5336

0.0122

289.87

−136.93

CA:Buprenorphine

−0.051 (0.0186)

177

−2.7413

0.0067

294.60

−139.30

CA:Naltrexone

−0.0757 (0.029)

177

−2.6136

0.0097

294.60

−139.30

Table 6

Repeated measures regression of selected CVS risk factors for log RA/CA

Opiates

Estimate (SE)

SE

df

t value

P value

All patients

SP

0.5976 (0.0947)

0.0947

625

6.3084

0.0000

Buprenorphine

−0.2318 (0.0408)

0.0408

625

−5.6869

0.0000

Naltrexone

−0.2873 (0.0579)

0.0579

625

−4.9605

0.0000

SP:Cigarettes

0.0051 (0.0024)

0.0024

625

2.1405

0.0327

Cigarettes:BMI

−0.0071 (0.0036)

0.0036

625

−1.9942

0.0466

Males

SP

0.5993 (0.1101)

0.1101

434

5.4434

0.0000

SP:Buprenorphine

−0.0499 (0.0108)

0.0108

434

−4.6199

0.0000

SP:Naltrexone

−0.0618 (0.0145)

0.0145

434

−4.2715

0.0000

Females

SP

1.2035 (0.1852)

0.1852

177

6.4967

0.0000

SP:Buprenorphine

−0.0262 (0.0131)

0.0131

177

−1.9983

0.0472

A total of 1,263 patients comprised of 866 (68.6 %) males and 397 females were studied in four groups as shown in Table 1. The baseline group was controls, and three treatment groups were treated each with buprenorphine, methadone or naltrexone. As shown in Table 1, these patients were studied on 2020 occasions. A total of 10,468 individual studies were performed, a mean of 5.18 studies per occasion. Repeated studies were excluded from the cross-sectional analysis. Table 1 also presents biometric, drug use and laboratory data. The control group is noted to be significantly younger than the treatment groups. The mean treatment dose of buprenorphine in these patients was 6.98 ± 0.21 mg (mean ± SEM), and of methadone was 63.00 ± 4.01 mg. Further drug use data from this cohort have been previously published [29]. Significant differences in drug use and some laboratory values are noted between the groups.

Log transformation of CA and RA was used in all statistical calculations as the Shapiro–Wilks “W” statistic was improved by the transformation of CA from 0.9305 to 0.9780, and of RA from 0.84725 to 0.9048 (in the cross-sectional dataset). Log transformation of the RA/CA ratio improves its normality compliance from W = 0.9302 to 0.9939.

Bivariate comparisons of cardiovascular parameters relating to vascular age, arterial stiffness, timing and pressure are presented in Supplementary Table 1. The operator index of all patients with accepted studies was noted to be 87–90 with no significant between-group differences. Significant differences in vascular age and arterial stiffness are noted between groups, but this must be interpreted with care in view of the significant between-group differences in chronological age (CA). For the same reason, adjustments for multiple testing have not been performed.

Table 1 also demonstrates significant differences between the groups in lifetime exposure to opiates. To control for this, a sub-analysis was performed in two groups of methadone patients (N = 60 and N = 22) where age, height, weight, BMI, heroin dose and duration and lifetime opiate use were matched against buprenorphine (N = 60)- and naltrexone (N = 22)-treated patients. The results of the bivariate analysis are shown in Supplementary Table 2. Significant differences in the vascular age, the RA-CA difference and the (log) RA/CA were found with buprenorphine, and the RA-CA difference and the (log) RA/CA ratio with naltrexone.

The between-group differences in CA were further addressed by multiple linear regression. Supplementary Table 3 presents the results of linear regression of the log of the vascular age against several variables known from the PWA literature to be important determinants of RA including CA, heart rate, height and also the treatment group. Patient’s treatment group is noted to be significant in comparison with these factors in the whole group overall, and in each sex separately.

Figure 1 presents various indices of vascular ageing by the treatment group. Similar figures are presented in Supplementary Figures 1–3 for arterial stiffness, timing and pressure. Figure 1 suggests greater differences between the four groups. These are quantified in Table 2 which presents the results of semi-logarithmic additive linear regression models of major central CVS parameters against chronological age and treatment type. As methadone is often said to be the “gold standard” of treatment [30], it has been used as the comparator agent, and so does not appear in these tables. Table 3 presents similar information for simple interactive final models between CA and treatment status. One will note some slight variation in the degrees of freedom related to missing data in the C_AP_HR75 and C_AGPH_HR75 groups. These missing data points were omitted in the analysis. Again methadone is the comparator group and so does not appear in the tables.

The matched groups referred to above were also assessed by multiple regression controlled for opiate dose and duration exposure. In an interactive model with CA, the age–buprenorphine (est. = 0.05425, t = −2.873, P = 0.00471; model Adj’d R2 = 0.03678, F = 28.35, df = 3,138, P = 2.4 × 10−14) and age–naltrexone interactions (est. = −0.0598, t = −2.277, P = 0.0243) were significant.

When the log (RA/CA) was regressed against the treatment group and the opiate exposure for all patients, the results shown in Supplementary Table 4 were obtained (model R2 = 0.348, F = 4.74, df = 12,1230, P = 1.3 × 10−7).

These data suggest that these patients’ vasculature appears to be ageing at different rates. Semi-log modelling of these various regression lines finds intercepts of 2.5625 (controls), 2.5365 (buprenorphine), 2.4891 (methadone) and 2.5518 (naltrexone) and slopes of 0.0275, 0.0290, 0.0321 and 0.0288 for the four groups, respectively. At the age of 60 years, these data predict vascular ages of 67.40, 72.03, 82.79 and 72.35, which are elevations of 4.63, 15.38 and 4.95 years above the control age, equivalent to 6.87, 22.84 and 7.35 %, respectively. Modelling calculations reveal that these data are progressive with CA, representing only 0.49, 1.99 and 1.67 % elevations above control at the age of 20 years.

Table 4 presents multiple regression output for a model with major cardiovascular risk factors included. The dependent variable was the log of the RA/CA ratio. The risk factors included were interactions between (log) brachial systolic blood pressure, daily tobacco consumption, (log) HDL, arcsinh(CRP), sex (when both sexes were considered) and pharmacotherapy group, together with additive terms in cholesterol, height and heart rate. As this study considers major cardiovascular risk factors in opiate dependence, it was limited to opiate-dependent patients only. In all patients considered together and in both sexes, interactions with treatment groups are significant after full adjustment for other established risk factors.

The numbers of patients in each group in the longitudinal study are given in Table 1. Figure 2 presents the results of following these four groups over log time up to 1904 days (5.21 years). Supplementary Figure 4 illustrates the same data over linear time. The lack of overlap between the standard error shaded regions suggests significant differences. Quantitation of these differences is shown in Table 5 which provides significant additive and interactive data for repeated measures mixed effects analysis by treatment group with age. The dependent variable is log(RA). Fixed effects are the chronologic age and the treatment group. Again the analysis is limited to opiate-dependents only. Table 6 presents a repeated measures mixed effects in opiate dependence for the log of the RA/CA ratio (to limit the very large number of interactions) against (log) CA, (log) systolic pressure, daily tobacco consumption, (log) BMI and treatment group. Once again, in all patients considered together, and in both sexes, interactions including treatment group cohort were significant.

Discussion

This study documents highly significant effects of different opiate dependence pharmacotherapy treatments (methadone, buprenorphine or naltrexone) on arterial stiffness and vascular age both as a factor and in interaction with chronological age. Significant differences were shown in age- and opiate exposure-matched groups with the RA–CA difference and the (log) RA/CA ratio. Significant interactions by treatment type persisted in fully adjusted multivariate models in both longitudinal and cross-sectional analyses, and in multivariate models adjusted for opiate exposure. These findings are therefore robust, highly provocative and compelling in their uniformity and diversity. This observational study used multiple regression models which correct for age and sex differences and identified highly statistically significant differences in arterial stiffness reflected in vascular age, central systolic pressure and subendocardial perfusion ratio by opiate treatment type. Of the three treatments, buprenorphine, methadone and naltrexone, methadone was uniformly associated with poorer arterial and vascular outcomes, both as a factor alone and in interaction with chronologic age and other established risk factors. In multiple regression limited to opiate-dependent patients, treatment type remained significant after adjustment for other established and novel risk factors including systolic pressure, tobacco consumption, cholesterol, HDL, CRP, height and heart rate. These data were equivalent to advancement in vascular age at a modelled CA of 60 years to 67.40, 72.3, 82.79 and 72.35 years or by 6.87, 22.84 and 7.35 %, respectively, compared to controls. Similar findings were reached in the longitudinal study when the (log) RA/CA ratio was regressed against systolic pressure, tobacco use, BMI and pharmacotherapy type.

These results appear to follow a consistent pattern of increasing arterial stiffness and increasing vascular age with an increasing level of opiate agonism. Methadone is a long-acting opiate agonist acting on the μ-, delta and κ-opioid receptors with a half-life of approximately 22 h in long-term users. It is commonly administered orally once a day for management of heroin dependence, with patients having almost 24-h opioid coverage and only experiencing no or mild withdrawal immediately prior to their next dose. Buprenorphine is a partial μ-receptor opiate agonist and an antagonist at κ-receptors allowing partial stimulation and blockade of these opioid receptors, respectively, which is similar to methadone delivered one daily, thereby providing 24-h coverage, with little withdrawal effects observed between doses [5]. Naltrexone is a long-acting (24+ h) full opiate antagonist that blocks both endogenous and exogenous opioid receptor stimulation. Hence, the pattern of increasing arterial stiffness and vascular age observed in the current study is associated with increasing levels of opiate receptor agonism. Shorter-acting opiates such as morphine, oxycodone and heroin are associated with several daily episodes of opiate withdrawal, which becomes a primary stimulus to re-administer them to relieve the aversive effects of opiate withdrawal. Since opiate withdrawal has been shown to be related to adverse hypertensive and other cardiovascular changes [31], exposing patients to unnecessary episodes of withdrawal is also not to be recommended.

As such, these findings have profound implications for the choice of pharmacotherapy agent for the management of opiate and heroin dependence, and the duration for which such treatment should be maintained. At present, methadone and buprenorphine are commonly recommended or practised for indefinite agonist maintenance over several decades [4, 5]. The present findings achieve even greater significance when it is recognized that since more than half of all patients die in Western nations from the consequences of cardiovascular disease, over half the effect of ageing on the human organism can be attributed to age-related cardiovascular degeneration [32]. Hence, the findings of pharmacotherapy-related accelerated central vascular stiffening documented in the present work are likely to carry much broader implications for the pathophysiology of the totality of the human organism.

Some discussion of the physiology of these various measures of arterial stiffness and their relationship with age is in order. The reader is referred to the Introduction for discussion of the Augmentation Pressure and its standardization by correction for pulse height as the Augmentation Index. Correction for pulse rate normalizes the impact of a rising heart rate to elevate stiffness indices. With age, the central diastolic pressure tends to fall and the central systolic pressure tends to rise. For this reason, their difference, known as the Central Pulse Height (C_PH), is an approximate measure of arterial stiffness. As one progresses distally in the arterial tree, the rising resistance tends to similarly elevate systolic and reduce diastolic pressures. This is known as the central-to-peripheral Pulse Pressure Amplification (PPAmpRatio). Because the central systolic pressure tends to rise with age, the amplification of this effect tends to be reduced with age, so that PPAmpRatio is negatively related to chronological age. The time to return to the aortic root of the reflected wave from the peripheral arteriolar resistance sites is a measure of arterial stiffness and is denoted as C_T1R. Measures of central and peripheral systolic blood pressure tend to rise with age. This is also reflected in the central mean pressure (C_MEANP) and central end systolic pressure (C_ESP). Other measures of pressure and timing reflect mean parameters produced from the cardiac cycle but are not directly related to vascular stiffness.

As noted in the “Results” section, use of the log transformation improves the normality assumptions of the RA and CA. Since the arterial tree becomes stiffer with age, the difference between RA and CA rises. This is the significance of the RA–CA difference presented in the Graphs and Tables. Use of the RA/CA ratio normalizes the effect of age directly in the dependent variable. Since there is such a wide range of patient ages, this is a very useful adjustment which directly facilitates data comparison, presentation and analysis. As noted, its normality is also significantly improved by log transformation. The operator index is a measure of the technical acceptability of a PWA study. Above 70 is regarded as satisfactory.

There is also literature examining the increased occurrence and premature appearance of many age-related pathologies in these patients [3, 15, 19]. Indeed, a Web Appendix (No. 6) of one study which investigated a very large cohort of heroin-dependent patients treated with methadone (and to a lesser extent buprenorphine) reported highly significant elevations of standardized mortality ratio for degenerative diseases of essentially every body system [3], with elevated rates of single organ and multi-organ system disease identified by another study [15]. Clearly, such changes would be theoretically consistent with opiate retardation of cellular and tissue growth [6, 25] via activation of the senescence locus on chromosome 9p21.3, identified by genome-wide association studies as being linked with cardiovascular diseases, diabetes and other degenerative disorders [8, 9, 10, 11]. It is interesting to note that the binding sites identified at this locus occur at a relative “gene desert”, and activities appear to correspond better with a long non-coding RNA as noted above [12, 13]. ANRIL has been suggested to act at the CDKN2A promoter site by activation by a promoter common to both it and CDKN2A, one activating transcription in the sense and one in the anti-sense direction; a γ-interferon-mediated mechanism; or by a cis-acting mechanism of interference with gene expression at a distance on the same chromosome [12, 33, 34]. Clearly, activation of the senescence locus would account for the generalized ageing activity of opiate agonists across all body systems. Hence, it is possible that the opiate dependence model might offer unique insights into the pathogenesis and development of coronary artery disease.

It is likely in the case of opiates that such changes would be exacerbated by their pro-apoptotic effect and by their immunostimulatory effect, since almost all lines of stem cells are potently inhibited by immune activity. Opiates also act via toll-like receptor 4 (TLR4, with downstream transduction through JAK-STAT signalling to NF-κB), mitogen-activated protein (MAP) kinase, phosphoinositol-3 kinase (PI3 K), transforming growth factor β (TGFβ), Akt-mTOR, via the c-Jun—Fos AP-1 system and gyanyl cyclase-dependent pathways which signal to nitric oxide synthase [30, 35, 36, 37]. Opiates signal to potassium and calcium receptors including those concerned with cardiac repolarization [16]. Opiates therefore have pleiotropic pharmacological effects. Hence, the effects of opiates to increase apoptosis, cellular senescence, reduce stem cell activity and stimulate diffuse immune activation are likely to compound each other. Furthermore, they are likely to involve all body tissues.

A further fascinating possibility suggested by the generalized pattern of ageing is that opiate agonist maintained patients may have shorter telomeres, or relatively more short telomeres, than controls. These shorter telomeres may then signal directly via RAP1 (repressor/activator protein-1), TERRA (telomere repeat containing RNA), P21 or other pathways to both NF-κB(nuclear factor kappa-B)-mediated immune hyperstimulation and the metabolic syndrome and insulin resistance as has recently been suggested [38, 39, 40].

As mentioned, opiate dependence is typically a long-term clinical problem characterized by multiple and recurrent period of relapse and retreatment over several years [4, 5]. To date, cumulative ill health and associated negative social consequences have not been considered when developing government or health policy and programmes to address opiate/heroin dependence or other long-term use of opiate-based drugs to treat other chronic medical conditions. Presumably, the multiple and recurrent period of relapse and retreatment of these conditions over several years has a compounding effect in terms of the cumulative deleterious effects on systemic health of patients managed under such regimes.

The present study has a number of shortcomings. First, the study was observation of non-randomized cohorts and validation of current study finding through a randomized prospective longitudinal study looking at endophenotypes of cardiovascular and general health is desirable. The drug history taken in these patients was transcribed in narrative form which does not readily lend itself to statistical analysis. Attention should be given to this in future iterations of such work. The use of bed-side “instant” urine drug tests and breathalysers would assist in the formalization of instantaneous drug use at the time of study. Alcohol in particular is a vasodilator and may falsely depress the arterial stiffness and measures of vascular age. Future replications could also measure other indices of arterial stiffness by more sophisticated methodologies. In terms of the generalizability of this study, no racial data were collected on which to base a comment. However, the size of the study, together with its consistency with the international literature cited, suggests that its effects may well be generalizable to other populations. Formal demonstration of this would require replication in pertinent populations. Finally, whilst a wide range of variables was included in the multiple regression analyses, it was not possible to control for unmeasured confounders such as diet, incarceration experience or infections. Consideration might be given to these in future iterations of such work.

The health implications of treating chronic disorders with opiate-based drugs, including drug dependence and drug abuse, are only beginning to be recognized. A major premise underlying medicine is that of ‘not doing harm’. Clearly, provision of any treatment which has major negative implications for other health variables needs to be weighed carefully against cost benefit. Further, given the widespread international desirability to minimize drug use and abuse, and improve the general health of populations, it would appear important to foster and encourage such research.

Given this, future studies should focus on cellular and molecular markers of ageing such as structural and functional studies of circulating stem cell populations, minimal telomere length, H2AγX activation, P16INK4A expression, cytokine activation, and deep genetic and epigenetic sequencing and study of the chromosome 9p21.3 senescence locus. Such studies may well improve our understanding of the basic pathophysiological processes underlying atherogenesis and the relationship of the senescence locus with the development of age-related clinical disease. As noted, the present study raises many mechanistic questions, many of which would be well suited to study in animal models. No such experimental studies are available to the author’s knowledge at the present time.

In addition to more detailed cardiovascular studies and arterial functional assessment techniques, the general implications of this work for whole of body ageing imply that prospective patient imaging could focus beyond the cardiovascular system, and so volumetric MRI’s of the hippocampus and bone mineral densitometry and immunological profiling would be valid further developments. Most importantly, together with the advent of newer treatments for opiate dependency such as buprenorphine, naltrexone implants and mixed agonist–antagonist dimers, such evidence should begin to foster a wider debate on the long-term burden on health and costs to patients and the community, of both opiate dependency and its treatment.

Notes

Conflict of interest

None.

Supplementary material

12012_2013_9204_MOESM1_ESM.ppt (451 kb)
Supplementary material 1 (PPT 451 kb)
12012_2013_9204_MOESM2_ESM.doc (120 kb)
Supplementary material 2 (DOC 119 kb)

References

  1. 1.
    Centers for Disease Control, Prevention (CDC). (2011). Vital signs: overdoses of prescription opioid pain relievers—United States, 1999–2008. Morbidity and Mortality Weekly Report, 60, 1487–1492.Google Scholar
  2. 2.
    Jamison, R. N., Serraillier, J., & Michna, E. (2011). Assessment and treatment of abuse risk in opioid prescribing for chronic pain. Pain Research and Treatment, 2011, 941808.PubMedCrossRefGoogle Scholar
  3. 3.
    Degenhardt, L., Randall, D., Hall, W., Law, M., Butler, T., & Burns, L. (2009). Mortality among clients of a state-wide opioid pharmacotherapy program over 20 years: Risk factors and lives saved. Drug and Alcohol Dependence, 105(1–2), 9–15.PubMedCrossRefGoogle Scholar
  4. 4.
    Kimber, J., Copeland, L., Hickman, M., Macleod, J., McKenzie, J., De Angelis, D., et al. (2010). Survival and cessation in injecting drug users: prospective observational study of outcomes and effect of opiate substitution treatment. BMJ (Clinical research ed), 341, c3172.CrossRefGoogle Scholar
  5. 5.
    Brunton, L. L., Lazo, J. S., & Parker, K. L. (Eds.). (2006). Goodman and Gilman’s the pharmacologic basis of therapeutics (11th ed., Vol. 1). New York: McGraw Hill.Google Scholar
  6. 6.
    McLaughlin, P. J., Zagon, I. S., & White, W. J. (1978). Perinatal methadone exposure in rats. Effects on body and organ development. Biology of the Neonate, 34(1–2), 48–54.PubMedGoogle Scholar
  7. 7.
    Cheng, F., McLaughlin, P. J., Verderame, M. F., & Zagon, I. S. (2009). The OGF-OGFr axis utilizes the p16INK4a and p21WAF1/CIP1 pathways to restrict normal cell proliferation. Molecular Biology of the Cell, 20(1), 319–327.PubMedCrossRefGoogle Scholar
  8. 8.
    Helgadottir, A., Thorleifsson, G., Manolescu, A., Gretarsdottir, S., Blondal, T., Jonasdottir, A., et al. (2007). A common variant on chromosome 9p21 affects the risk of myocardial infarction. Science, 316(5830), 1491–1493.PubMedCrossRefGoogle Scholar
  9. 9.
    McPherson, R., Pertsemlidis, A., Kavaslar, N., Stewart, A., Roberts, R., Cox, D. R., et al. (2007). A common allele on chromosome 9 associated with coronary heart disease. Science, 316(5830), 1488–1491.PubMedCrossRefGoogle Scholar
  10. 10.
    Feero, W. G., Guttmacher, A. E., & McCarthy, M. I. (2010). Genomics, Type 2 diabetes, and obesity. New England Journal of Medicine, 363(24), 2339–2350.CrossRefGoogle Scholar
  11. 11.
    O’Donnell, C. J., & Nabel, E. G. (2011). Genomics of cardiovascular disease. New England Journal of Medicine, 365(22), 2098–2109.PubMedCrossRefGoogle Scholar
  12. 12.
    Pasmant, E., Sabbagh, A., Vidaud, M., & Bieche, I. (2010). ANRIL, a long, noncoding RNA, is an unexpected major hotspot in GWAS. The FASEB Journal, 25(2), 444–448.CrossRefGoogle Scholar
  13. 13.
    Holdt, L. M., Beutner, F., Scholz, M., Gielen, S., Gabel, G., Bergert, H., et al. (2010). ANRIL expression is associated with atherosclerosis risk at chromosome 9p21. Arteriosclerosis, Thrombosis, and Vascular Biology, 30(3), 620–627.PubMedCrossRefGoogle Scholar
  14. 14.
    Khademi, H., Malekzadeh, R., Pourshams, A., Jafari, E., Salahi, R., Semnani, S., et al. (2012). Opium use and mortality in Golestan Cohort Study: prospective cohort study of 50, 000 adults in Iran. BMJ (Clinical research ed), 344, e2502.CrossRefGoogle Scholar
  15. 15.
    Darke, S., Kaye, S., & Duflou, J. (2006). Systemic disease among cases of fatal opioid toxicity. Addiction, 101(9), 1299–1305.PubMedCrossRefGoogle Scholar
  16. 16.
    Sadeghian, S., Darvish, S., Davoodi, G., Salarifar, M., Mahmoodian, M., Fallah, N., et al. (2007). The association of opium with coronary artery disease. European Journal of Cardiovascular Prevention & Rehabilitation, 14(5), 715–717.CrossRefGoogle Scholar
  17. 17.
    Sadeghian, S., Dowlatshahi, S., Karimi, A., & Tazik, M. (2011). Epidemiology of opium use in 4398 patients admitted for coronary artery bypass graft in Tehran Heart Center. The Journal of Cardiovascular Surgery (Torino), 52(1), 140–141.Google Scholar
  18. 18.
    Claude Bernard, M. (1877). Lecons Sur Le Diabete et la Glycogenese Animale (Vol. 1). Paris: Cours de Medecine.Google Scholar
  19. 19.
    Rosen, D., Smith, M. L., & Reynolds, C. F., 3rd. (2008). The prevalence of mental and physical health disorders among older methadone patients. American Journal of Geriatric Psychiatry, 16(6), 488–497.PubMedCrossRefGoogle Scholar
  20. 20.
    Cooper, O. B., Brown, T. T., & Dobs, A. S. (2003). Opiate drug use: a potential contributor to the endocrine and metabolic complications in human immunodeficiency virus disease. Clinical Infectious Diseases, 37(Suppl 2), S132–S136.PubMedCrossRefGoogle Scholar
  21. 21.
    Kolarzyk, E., Pach, D., Wojtowicz, B., Szpanowska-Wohn, A., & Szurkowska, M. (2005). Nutritional status of the opiate dependent persons after 4 years of methadone maintenance treatment. Przeglad Lekarski, 62(6), 373–377.PubMedGoogle Scholar
  22. 22.
    Darke, S., Duflou, J., & Torok, M. (2010). The comparative toxicology and major organ pathology of fatal methadone and heroin toxicity cases. Drug and Alcohol Dependence, 106(1), 1–6.PubMedCrossRefGoogle Scholar
  23. 23.
    Reece, A. S. (2009). Dental health in addiction. Australian Dental Journal, 54(2), 185–186.PubMedCrossRefGoogle Scholar
  24. 24.
    Friedewald, V. E., Kornman, K. S., Beck, J. D., Genco, R., Goldfine, A., Libby, P., et al. (2009). The American Journal of Cardiology and Journal of Periodontology editors’ consensus: periodontitis and atherosclerotic cardiovascular disease. Journal of Periodontology, 80(7), 1021–1032.PubMedCrossRefGoogle Scholar
  25. 25.
    Zagon, I. S., Verderame, M. F., & McLaughlin, P. J. (2002). The biology of the opioid growth factor receptor (OGFr). Brain Research Reviews, 38(3), 351–376.PubMedCrossRefGoogle Scholar
  26. 26.
    Nichols, W. W., Denardo, S. J., Wilkinson, I. B., McEniery, C. M., Cockcroft, J., & O’Rourke, M. F. (2008). Effects of arterial stiffness, pulse wave velocity, and wave reflections on the central aortic pressure waveform. The Journal of Clinical Hypertension (Greenwich), 10(4), 295–303.CrossRefGoogle Scholar
  27. 27.
    Laurent, S., Boutouyrie, P., Asmar, R., Gautier, I., Laloux, B., Guize, L., et al. (2001). Aortic stiffness is an independent predictor of all-cause and cardiovascular mortality in hypertensive patients. Hypertension, 37(5), 1236–1241.PubMedCrossRefGoogle Scholar
  28. 28.
    AtCor Medical Corporation. (2006). SphygmoCor pulse wave analysis system model SCOR-Px. PP1–63 (1st ed., Vol. 1). Sydney: Atcor Medical.Google Scholar
  29. 29.
    Reece, A. S. (2007). Psychosocial and treatment correlates of opiate free success in a clinical review of a naltrexone implant program. Substance Abuse Treatment, Prevention, and Policy, 2, 35–49.PubMedCrossRefGoogle Scholar
  30. 30.
    Hulse, G., White, J., & Cape, G. (Eds.). (2002). Management of drug and alcohol problems (Vol. 1). Sydney: Oxford University Press.Google Scholar
  31. 31.
    Kienbaum, P., Thurauf, N., Michel, M. C., Scherbaum, N., Gastpar, M., & Peters, J. (1998). Profound increase in epinephrine concentration in plasma and cardiovascular stimulation after mu-opioid receptor blockade in opioid-addicted patients during barbiturate-induced anesthesia for acute detoxification. Anesthesiology, 88(5), 1154–1161.PubMedCrossRefGoogle Scholar
  32. 32.
    Chien, K. R., & Karsenty, G. (2005). Longevity and lineages: toward the integrative biology of degenerative diseases in heart, muscle, and bone. Cell, 120(4), 533–544.PubMedCrossRefGoogle Scholar
  33. 33.
    Visel, A., Zhu, Y., May, D., Afzal, V., Gong, E., Attanasio, C., et al. (2010). Targeted deletion of the 9p21 non-coding coronary artery disease risk interval in mice. Nature, 464(7287), 409–412.PubMedCrossRefGoogle Scholar
  34. 34.
    Harismendy, O., Notani, D., Song, X., Rahim, N. G., Tanasa, B., Heintzman, N., et al. (2011). 9p21 DNA variants associated with coronary artery disease impair interferon-gamma signalling response. Nature, 470(7333), 264–268.PubMedCrossRefGoogle Scholar
  35. 35.
    Hutchinson, M. R., Zhang, Y., Shridhar, M., Evans, J. H., Buchanan, M. M., Zhao, T. X., et al. (2010). Evidence that opioids may have toll-like receptor 4 and MD-2 effects. Brain, Behavior, and Immunity, 24(1), 83–95.PubMedCrossRefGoogle Scholar
  36. 36.
    Yin, D., Woodruff, M., Zhang, Y., Whaley, S., Miao, J., Ferslew, K., et al. (2006). Morphine promotes Jurkat cell apoptosis through pro-apoptotic FADD/P53 and anti-apoptotic PI3 K/Akt/NF-kappaB pathways. Journal of Neuroimmunology, 174(1–2), 101–107.PubMedCrossRefGoogle Scholar
  37. 37.
    Grzanna R., & Brown R. M. (1993). NIDA research monograph 125: Activation of intermediate early genes by drugs of abuse. In: National Institutes of Drug Abuse (Ed.) (Vol. 125, pp. 1–220). Bethesda, Maryland: NIH.Google Scholar
  38. 38.
    Varela, E., & Blasco, M. A. (2010). 2009 Nobel Prize in physiology or medicine: telomeres and telomerase. Oncogene, 29(11), 1561–1565.PubMedCrossRefGoogle Scholar
  39. 39.
    Martinez, P., & Blasco, M. A. (2011). Telomeric and extra-telomeric roles for telomerase and the telomere-binding proteins. Nature Reviews Cancer, 11(3), 161–176.PubMedCrossRefGoogle Scholar
  40. 40.
    Martinez, P., Thanasoula, M., Carlos, A. R., Gomez-Lopez, G., Tejera, A. M., Schoeftner, S., et al. (2010). Mammalian Rap1 controls telomere function and gene expression through binding to telomeric and extratelomeric sites. Nature Cell Biology, 12(8), 768–780.PubMedCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  1. 1.School of Psychiatry and Clinical NeurosciencesUniversity of Western AustraliaHighgate Hill, BrisbaneAustralia

Personalised recommendations