Cardiovascular Changes Associated With Venlafaxine in the Treatment of Late-Life Depression

Transcription

1 Cardiovascular Changes Associated With Venlafaxine in the Treatment of Late-Life Depression Ellyn M. Johnson, Ellen Whyte, M.D., Benoit H. Mulsant, M.D., Bruce G. Pollock, M.D., Ph.D., Elizabeth Weber, C.R.N.P., Amy E. Begley, M.A., Charles F. Reynolds, M.D. Background: Potential cardiovascular side effects from venlafaxine-xr must be considered when prescribing this medication, especially in geriatric patients, who often present with comorbid medical conditions. Methods: Participants age 60 and older with a Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition diagnosis of a major depressive episode without psychotic features were treated openly with venlafaxine-xr for 12 weeks during which venlafaxine-xr was titrated based on tolerability and response according to predefined guidelines. Sitting and standing blood pressures and heart rates were measured. A 12-lead electrocardiogram was obtained at baseline and at week 12. Results: Sixty-two participants started treatment; 59 completed at least two weeks of the 12-week study. The mean final dose of venlafaxine-xr was mg/day (standard deviation: 72.2). Twentyfour percent (95% confidence interval [CI]: 7.3% 40.7%) of initially normotensive participants and 54% (95% CI: 34.3% 74%) of those with preexisting hypertension experienced an increase in blood pressure. Twenty-nine percent (95% CI: 14.6% 43.4%) of participants developed orthostatic hypotension. Two participants experienced a clinically significant increase in QTc interval. One participant reported new-onset mild dizziness, whereas four participants reported new-onset tachycardia or palpitation. Overall, 17 unique participants (28.8%; 95% CI: 17.3% 40.4%) experienced a new-onset cardiovascular problem, potentially related to the study medication. Conclusion: Overall, venlafaxine-xr was well tolerated. However, similar to previous reports, venlafaxine-xr was associated with some undesirable cardiovascular effects in some of the participants. Systematic monitoring of cardiovascular parameters during treatment with venlafaxine-xr should be strongly recommended, especially in the elderly. (Am J Geriatr Psychiatry 2006; 14: ) Received June 10, 2005; revised December 19, 2005; accepted January 5, From the Western Psychiatric Institute and Clinic, Department of Psychiatry (EMJ, EW, BHM, BGP, EW, AEB, CFR), University of Pittsburgh School of Medicine, Pittsburgh, PA; the Center for Addiction and Mental Health and Department of Psychiatry (BHM), University of Toronto, Ontario, Canada; the Rotman Research Institute (BGP), Baycrest Center for Geriatric Care, University of Toronto, Ontario, Canada; and the Department of Neurology (CFR), University of Pittsburgh School of Medicine, Pittsburgh, PA. Send correspondence and reprint requests to Dr. Benoit H. Mulsant, WPIC (E837), 3811 O Hara St., Pittsburgh, PA benoit_mulsant@camh.net 2006 American Association for Geriatric Psychiatry 796

2 Johnson et al. Late-life depression is a significant health problem affecting millions of older Americans every year. 1 These older persons require active treatment to relieve the suffering and the cognitive and functional impairment associated with depression and to prevent the associated physical morbidity and mortality. 2 When they are treated, most older Americans with depression receive a selective serotonin reuptake inhibitor (SSRI). 3 However, between one-fourth and one-half do not respond to a first-line SSRI and require alternative treatment. 4,5 As a result of its dual action on both the serotonin and norepinephrine reuptake, venlafaxine is often used as a second-line antidepressant in younger and older patients. 3,4 This dual action has been hypothesized to be associated with a higher rate of remission than treatment with a SSRI in younger patients with depression. 6 However, this potential higher efficacy has not been observed in older patients, 7,8 and in frail nursing home residents, venlafaxine has been found to be associated with a higher rate of adverse effects. 7 Venlafaxine has been reported to induce hypertension in 10% 20% of younger or older patients who take the higher doses that have been associated with dual reuptake inhibition. 9 In overdose, venlafaxine has been reported to cause tachycardia, paradoxic hypotension and prolonged QRS and QTc intervals. 10,11 CYP 450 2D6 poor metabolizers have also been shown to be at risk for cardiac symptoms, including severe dizziness, palpitations, and syncope. 12 Thus, we conducted a prospective study to evaluate the presence and extent of cardiovascular changes in a group of older patients treated with venlafaxine. We hypothesized that venlafaxine would be associated with an increase in QRS and QTc intervals, orthostatic hypotension, or systolic hypertension in a subgroup of patients. METHODS Participants were community-dwelling elderly who were recruited from January 2001 to May 2004 among all patients evaluated in the Late-Life Depression Program of Western Psychiatric Institute and Clinic at the University of Pittsburgh. The inclusion and exclusion criteria were designed to be as broad as possible so that the study group would closely represent older depressed patients who receive venlafaxine for the treatment of depression in clinical settings. Participants were required to meet the following criteria: age 60 and older; Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) diagnosis of a major depressive episode without psychotic features; 17-item Hamilton Rating Scale for Depression (HRSD) 13 score of 11 or above for participants who were currently treated with an antidepressant (i.e., absence of remission) or 15 or above for those who had not received any antidepressant medication during this episode (i.e., fully symptomatic depression); absence of suicidal intent; Mini-Mental State Examination (MMSE) 14 score of 15 or above; no history of meeting DSM IV criteria for mania, schizophrenia, or other psychotic disorder; no history of receiving venlafaxine immediate or extended release (XR) for six or more weeks at 300 mg/day during the index depressive episode; no unstable physical condition or absolute contraindication to treatment with either venlafaxine-xr or venlafaxine immediate-release formulation (i.e., QTc 480 msec, untreated or uncontrolled hypertension, current hyponatremia); and no myocardial infarction, unstable angina, palpitation, or cardiac syncope within three months before entering the study. All participants provided written informed consent according to University Institutional Review Board procedures. The comprehensive evaluation included administration of the Structured Clinical Interview for Axis-I DSM IV Disorders (SCID-IV), 15 a semistructured version of the 17-item HRSD, 13 the Cumulative Illness Rating Scale adapted for Geriatrics (CIRS-G) 16 from which the total and count scores and a cardiovascular score representing the sum of heart and vascular disease items (questions 1 and 2) were derived, the UKU side-effects scale (UKU), 17 and the MMSE. After a washout of all psychotropic medications, except for lorazepam that was allowed throughout the study if needed for severe anxiety or insomnia, participants were started on 37.5 mg venlafaxine XR every morning. Venlafaxine XR was then titrated upward by 37.5 or 75 mg each week until a target dose of 150 mg/day was reached. In participants whose HRSD was still 15 or above after four weeks of treatment, or 11 or above after six or eight weeks of treatment, the dose was increased by mg every other week up to a maximum dose of 300 mg/day. Doses were adjusted downward in the presence of significant side effects. This acute treatment study lasted 12 weeks. Par- 797

3 Cardiovascular Changes Associated With Venlafaxine ticipants were seen weekly for four weeks and then every other week for the last eight weeks of the study. At each visit, participants were assessed with the HRSD and UKU. At baseline, sitting and standing blood pressure and heart rate were assessed in 47 participants and lying and standing blood pressure and heart rate were assessed in 10 participants. Because the blood pressure and heart rate changes of these two groups of participants did not differ statistically (data not shown), the two groups were pooled for all analyses. At each follow-up visit, blood pressure and heart rate were measured in the sitting position after resting for 5 minutes and then 1 minute after standing. Blood pressure was measured using the Omron Automatic Blood Pressure Monitor (Omron Healthcare Inc., Bannockburn, IL); a large cuff was used if the arm circumference was greater than 33 cm. A 12-lead electrocardiogram (EKG) was obtained while the patient was reclining at baseline and after 12 weeks of treatment (or at termination for participants who did not complete the protocol). QTc interval was calculated using the Bazett correction (QT interval divided by the square root of 60/heart rate). 18 For patients who discontinued study medication before completing 12 weeks of acute treatment, the reasons for dropping out were determined and classified according to established rules. During the enrollment period, 377 outpatients were screened at the Late-Life Depression Program, 109 were invited to participate, 67 signed informed consent, and 62 initiated treatment with venlafaxine XR (i.e., they took at least one dose). Three participants were excluded from this analysis because they withdrew from the study within the first 10 days and lacked follow-up data. Baseline and follow-up data were available for 59 participants who constitute the study group for the analyses. For these analyses, hypertension was defined as sitting or lying systolic blood pressure above 140 mm Hg or a diastolic blood pressure above 90 mm Hg. Orthostatic hypotension was defined as either a drop in systolic blood pressure of 20 mm Hg or more, a drop in diastolic blood pressure of 10 mm Hg or more, or an increase in heart rate of 20 beats/min or more after one minute on standing from a sitting or lying position. Descriptive statistics were calculated to characterize the participants. EKGs were characterized by heart rate, PR interval, QRS interval, and QTc interval at baseline and week 12. Blood pressure, heart rate, and orthostatic blood pressure/heart rate are reported for baseline and week 12. In addition, the change in these parameters over the 12-week study is reported. The UKU was used to measure side effect burden at baseline and week 12. Total score is presented as well as item scores for orthostatic dizziness (item 28) and palpitations/tachycardia (item 29) because these items represent problems of interest. RESULTS Participants Characteristics The 59 participants had a mean (SD) age of 74.5 (7.5), 67.8% were female, 93.2% were white, with a mean of 13.7 (2.5) years of education (N 58), MMSE of 28.1 (2.1) (N 57), and baseline HRSD of 18.8 (3.7). At baseline, mean (SD) CIRS-G total score was 11.1 (4.7), CIRS-G count was 6.4 (2.2), and cardiovascular subscale score was 2.3 (1.8), reflecting a study group with a moderate physical burden and mild to moderate cardiac disease. Participants took venlafaxine-xr in the study for a mean (SD) of 11.3 (2.3) weeks (median: 11.9, range: ) with 53 (89.8%) participants completing the 12-week study. They were prescribed a mean (SD) maximum dose of venlafaxine-xr (61.4) mg/day (median: 225, range: ) and a final dose mean (SD) of (72.2) mg/day (median: 187.5, range: ). Of the six participants who prematurely left the study, 66% had hypertension at baseline. None demonstrated orthostatic blood pressure changes at baseline. The study dropouts did not differ from study completers on the CIRS-G. Medication Tolerability Overall, 17 unique participants (28.8%; 95% CI: 17.3% 40.4%) experienced a new-onset cardiovascular symptom potentially related to the study medication. One participant stopped taking the study medication after 10 weeks as a result of orthostatic hypotension, whereas another participant stopped the study medication at week 12 as a result of possible study medication-induced headaches. No patient discontinued the study as a result of EKG changes or onset of unmanageable hypertension. The total mean (SD) UKU score decreased from 15.2 (5.5, N 58) at 798

4 Johnson et al. baseline to 10.0 (3.9, N 54) at week 12 (t -7.21, df 49, p ). Of the 30 participants who did not endorse dizziness on the UKU (item 28) at baseline, eight reported mild dizziness at week 12. Of the 21 participants who endorsed mild dizziness at baseline, only one experienced a worsening of this symptom with treatment, whereas the remainder either experienced no change in or a resolution of this symptom by week 12. Four participants reported the new onset of palpitations or tachycardia (i.e., a score of 1 or 2 on the UKU palpitations/tachycardia item 29.) The three participants who endorsed palpitations/tachycardia at baseline reported symptom resolution by week 12. Week 12 UKU total score did not correlate with study medication dose (r 0.06, p 0.68, N 58), participant age (r 0.12, p 0.36, N 58), CIRS-G total score (r 0.15, p 0.25, N 58), or CIRS-G cardiovascular score (r 0.24, p 0.07, N 58). Electrocardiogram Changes Baseline EKGs were available for 55 participants and follow-up EKGs for 42. A total of 40 participants had both baseline and follow-up EKGs. Table 1 summarizes the EKG measures at baseline and at follow up. A total of seven unique participants (17.5%; 95% CI: 5.7% 29.3%) experienced a change in their EKG. Considering a QTc interval longer than or equal to 440 msec as clinically significant, one participant (2.5%) experienced an increase in QTc interval from 403 to 456 and another one (2.5%) from 433 to 469. Three participants (7.5%) developed nonspecific ST TABLE 1. Electrocardiogram Data Before and at Completion of Treatment With Venlafaxine-XR a Baseline Week 12 t-statistic, df (N 55) (N 42) (p value) Heart rate 75.1 (12.5) 78.6 (14.2) 2.00, 39 (beats/min) [50 107] [54 122] (0.052) PR interval (23.5) b (28.9) c 0.24, 37 (msec) [ ] [ ] (0.82) QRS 88.7 (14.7) 87.7 (16.0) 1.44, 39 [68 160] [70 174] (0.16) QTc (21.3) (21.2) 0.25, 39 [ ] [ ] (0.81) a All results are expressed as mean (standard deviation) [range]. b N 52; PR interval could not be measured on three participants as a result of poor quality of electrocardiogram, atrial fibrillation, or presence of a pacemaker. c N 40; PR interval could not be measured on two participants as a result of atrial fibrillation or presence of a pacemaker. wave changes, one participant developed premature ventricular contractions (PVCs), and one participant developed both nonspecific ST wave changes and PVCs. None of the participants demonstrated a new onset of a conduction defect, an increase in PR interval beyond 200 msec, or an increase in QRS interval beyond 120 msec. As assessed by the UKU, two of the seven unique participants with EKG changes reported new-onset dizziness, whereas none of these seven participants reported new onset or worsening of palpitations. The two participants with new-onset dizziness both demonstrated an increase in QTc interval 440msec. As a group, study completers experienced a statistically significant 3.5 beat/ min (bpm) increase in heart rate (mean [SD]: 75.1 bpm [12.5] at baseline versus 78.6 bpm [14.2] at week 12; t 2.00, df 39, p 0.052) over the 12-week study. No other EKG variable demonstrated statistically significant change over time. Hypertension Thirty-three of 59 (55.9%) participants were taking antihypertensive medications at baseline. Baseline and follow-up blood pressure readings and heart rates were available for 49 participants (Table 2). Twenty-five participants were normotensive at baseline (i.e., their systolic blood pressure was 130 or below and their diastolic blood pressure was 90 or below). Nine (36%; 95% CI: 17.2% 54.8%) of these initially normotensive individuals had at least two blood pressure readings in the hypertensive range over the course of the 12-week study. Using this definition of hypertension, the mean (SD) time to develop hypertension was 4.2 (3.6) weeks with a median of three weeks and a range of 1 12 weeks. Of these nine participants, four were not on antihypertensive medications at either baseline or week 12; two were receiving antihypertensive medications at baseline, but not at week 12; three had a change in their antihypertensive medication regimen during the study. Sixteen (64%; 95% CI: 45.2% 82.8%) of the 25 initially normotensive participants remained normotensive during the study. Of these 16 participants, two were receiving antihypertensive medications at baseline, but not at week 12, one started an antihypertensive medication during the study, and one had a change in antihypertensive medication regimen during the study. All participants with elevated 799

5 Cardiovascular Changes Associated With Venlafaxine TABLE 2. Cardiovascular Data Before and at Completion of Treatment With Venlafaxine-XR a Baseline Week 12 b t-statistic, df (N 57) (N 55) (p value) Sitting systolic blood pressure (16.5) (17.4) 0.01, 48 (0.99) [138] [133] Sitting diastolic blood pressure 76.2 (10.9) 78.3 (9.0) 1.18, 48 (0.24) [76] [80] Sitting heart rate 75.1 (10.1) 77.9 (12.8) 1.44, 48 (0.16) [72] [76] Standing systolic blood pressure (21.9) (15.6) 2.32, 48 (0.025) [133] [130] Standing diastolic blood pressure 76.1 (13.8) 76.9 (10.4) 0.07, 48 (0.95) [74] [78] Standing heart rate 79.6 (10.1) c 82.9 (13.7) 1.11, 45 (0.27) [80] [84] a All results are expressed as mean (standard deviation); range [median]. b Data gathered at week 12 or at time of participant s termination. Five participants terminated early: one participant at week 2, two participants at week 4, and two participants at week 6. c N 53. blood pressure readings were strongly encouraged to seek treatment through their primary care provider. No statistically significant increase in blood pressure was observed. Of the 24 participants who met criteria for hypertension at baseline, 13 still met the criteria for hypertension at week 12. These 13 participants demonstrated a mean (SD) change of 1.9 (14.1; range: 16 to 25, median: 2) mm Hg in sitting systolic and 0.38 (6.8; range: 10 to 13, median: 1) mm Hg in sitting diastolic blood pressure. Three of these 13 participants had their antihypertensive medication adjusted during the course of the 12-week study. Orthostatic Hypotension Baseline and follow-up sitting/lying and standing blood pressures and heart rates were available for 47 participants (Table 2). Of the 38 participants who were not orthostatic at baseline, 19 (50.0%; 95% CI: 34.1% 65.9%) met criteria for orthostatic hypotension at least twice over the course of the 12-week study. Two of these participants with new-onset orthostasis also developed new-onset hypertension during the study. The mean (SD) time to onset of orthostatic hypotension was 3.3 (2.3) weeks with a median of three weeks and a range of 1 10 weeks. Of the nine participants who initially met criteria for orthostasis, six (67%) met criteria for orthostasis at least twice over the course of the 12-week study. Over the 12-week study, the study completers demonstrated a statistically significant mean decrease in standing systolic blood pressure of 7.2 mm Hg (mean [SD]: baseline [21.9] versus week [15.6]; t 2.32, df 48, p 0.025). Other changes in blood pressure variables were not statistically significant. All subjects with symptomatic orthostatic changes were instructed to increase their fluid intake and to change position slowly. DISCUSSION Among 59 older patients treated with up to 300 mg/day of venlafaxine-xr in a 12-week study, two (4.8%) experienced a clinically significant increase in QTc interval, nine (36%) initially normotensive participants demonstrated two or more blood pressure readings in the hypertensive range over the 12-week study, and 19 (50%) experienced new-onset orthostatic hypotension. Overall, 17 (28.8%) unique participants experienced a potentially clinically significant cardiovascular adverse effect, but only one study participant discontinued venlafaxine-xr as a result of these changes. A statistically significant increase in heart rate (as measured by EKG) and decrease in standing blood pressure were observed. The later finding likely reflects the large number of participants who developed orthostatic blood pressure changes over this 12-week study. We did not identify any participant characteristics, which could identify individuals at increase risk for these cardiovascular changes. 800

6 Johnson et al. The strengths of this study include a close prospective monitoring of blood pressure, heart rate, EKG, and potential side effects in older outpatients with major depression. Potential limitations include the relatively small number of participants preventing us from detecting rare but more serious events and the absence of a control group preventing us from making definite causal attribution. In addition, we attempted to actively manage emergent adverse effects. This may have obfuscated the relationship between cardiovascular changes and symptoms endorsed by the participants. Nevertheless, our findings are consistent with previous reports of cardiovascular problems or frank cardiotoxicity in some younger and older patients treated with venlafaxine or venlafaxine-xr. 7,19 Increased attention is being paid to the liability of psychotropic medications to increase the QTc interval. The QTc interval represents the period of ventricular repolarization. Slow repolarization (e.g., prolonged QTc interval) is associated with an increased risk of potentially lethal arrhythmia. 20 Prolonged QTc is of particular concern in the elderly who already exhibit a high prevalence of cardiac pathology. Our observation of a clinically significant prolongation of QTc in two (4.8%) of our older participants treated with venlafaxine-xr is consistent with earlier data showing the occurrence of QTc prolongation with venlafaxine. 11 Hypertension is a common diagnosis among the elderly and a major risk factor for cardiovascular and cerebrovascular disease. Venlafaxine-XR s inhibition of norepinephrine reuptake can potentiate sympathetic vasoconstriction in a dosage-dependent fashion 21 and can lead to increased heart rate and blood pressure. 19 Over one-third of our participants who had normal blood pressure at baseline demonstrated at least two elevated blood pressure readings over the course of this 12-week study. Orthostasis is also a serious concern in the elderly as a result of the high risk of falls and related injuries. Because of its lack of affinity for 1 -adrenergic receptors, venlafaxine is usually not thought to be associated with orthostasis. 22 However, venlafaxine has been linked to hypotension, syncope, palpitations, and profound dizziness. 10,12 Indeed, nearly one-half of our participants developed new-onset orthostatic hypotension during treatment with venlafaxine-xr. This may contribute to the poor tolerability of venlafaxine in the very old. 7 The growing number of older Americans being diagnosed with, and treated for, depression 23 demands safe and effective medications to assist in a timely recovery. Dual-action antidepressants may have a special role in elderly patients with treatmentresistant depression. 4 However, drugs that affect the noradrenergic system can lead to potentially serious cardiovascular adverse effects. In this study, the absence of clinical symptoms reported by our participants, even in the presence of potentially significant cardiovascular changes, emphasizes the importance of this regular and careful monitoring instead of relying on subjective symptom reports. This is particularly important in patients with either phenotypic or genetic inhibition of CYP2D6 associated with decreased drug clearance and therefore at higher risk of cardiotoxicity. 24 For clinicians, the literature and our findings do not suggest that use of this medication should be avoided. Instead, the literature and our findings suggest the need for systematic monitoring of orthostatic blood pressure, heart rate, and EKG in elders treated with venlafaxine. The risk benefit ratio of every pharmacologic intervention must be assessed for each individual patient. If adverse effects are detected, they need to be managed aggressively to allow the patient to benefit from the pharmacologic treatment. For researchers, our findings emphasize the public health importance of prospective drug studies in elderly patients focused on the safety concerns specific to this vulnerable population. 7 References 1. Mulsant BH, Ganguli M: Epidemiology and diagnosis of depression in late life. J Clin Psychiatry 1999; 60 (suppl 20): Charney DS, Reynolds CF III, Lewis L, et al: Depression and Bipolar Support Alliance consensus statement on the unmet needs in diagnosis and treatment of mood disorders in late life. Arch Gen Psychiatry 2003; 60: Mulsant BH, Alexopoulos GS, Reynolds CF, et al: Pharmacological treatment of depression in older primary care patients: the PROS- PECT algorithm. Int J Geriatr Psychiatry 2001; 16: Whyte EM, Basinski J, Farhi P, et al: Geriatric depression treatment in nonresponders to selective serotonin reuptake inhibitors. J Clin Psychiatry 2004; 65:

7 Cardiovascular Changes Associated With Venlafaxine 5. Mulsant BH, Pollock BG, Nebes R, et al: A 12-week study doubleblind, randomized comparison of notriptyline and paroxetine in older depressed inpatients and outpatients. Am J Geriatr Psychiatry 2001; 9: Thase ME, Entsuah R, Cantillon M, et al: Relative antidepressant efficacy of venlafaxine and SSRIs: sex-age interactions. J Women Health 2001; 14: Oslin DW, Ten Have TR, Streim JE, et al: Probing the safety of medications in the frail elderly: evidence from a randomized clinical trial of sertraline and venlafaxine in depressed nursing home residents. J Clin Psychiatry 2003; 64: Allard P, Gram L, Timdahl K, et al: Efficacy and tolerability of venlafaxine in geriatric outpatients with major depression: a double-blind, randomised 6-month comparative trial with citalopram. Int J Geriatr Psychiatry 2004; 19: Venlafaxine package insert. Wyeth Pharmaceuticals. Philadelphia, PA Kokan L, Dart R: Life-threatening hypotension from venlafaxine overdose. Ann Emerg Med 1996;27: Peano C, Leikin JB, Hanashiro PK: Seizures, ventricular tachycardia, and rhabdomyolysis as a result of ingestion of venlafaxine and lamotrigine. Ann Emerg Med 1997; 30: Lessard E, Yessine MA, Hamelin BA, et al: Influence of CYP2D6 activity on the disposition and cardiovascular toxicity of the antidepressant agent venlafaxine in humans. Pharmacogenetics 1999; 9: Hamilton M: A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: Folstein MF, Folstein SE, McHugh PR: Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: Structured Clinical Interview for Axis I DSM IV Disorders Patient Edition (SCID-I/P). New York, Biometrics Research Department, NY State Psychiatric Institute, Miller MD, Paradis CF, Houck PR, et al: Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res 1992; 41: Lingjaerde O, Ahlfors UG, Bech P, et al: The UKU side effect rating scale. A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side effects in neuroleptictreated patients. Acta Psychiatr Scand Suppl 1987; 334: Miller MD, Curtiss EI, Marino L, et al: Long-term ECG changes in depressed elderly patients treated with nortriptyline. A doubleblind, randomized, placebo-controlled evaluation. Am J Geriatr Psychiatry 1998; 6: Thase ME: Effects of venlafaxine on blood pressure: a metaanalysis of original data from 3744 depressed patients. J Clin Psychiatry 1998; 59: Haddad PM, Anderson IM: Antipsychotic-related QTc prolongation, torsade de pointes and sudden death. Drugs 2002; 62: Abdelmawla AH, Langley RW, Szabadi E, et al: Comparison of the effects of venlafaxine, desipramine, and paroxetine on noradrenaline- and methoxamine-evoked constriction of the dorsal hand vein. Br J Clin Pharmacol 1999; 48: Feighner JP: Cardiovascular safety in depressed patients: focus on venlafaxine. J Clin Psychiatry 1995; 56: Harman JS, Mulsant BH, Kelleher KJ, et al: Narrowing the gap in treatment of depression. Int J Psychiatry Med 2001; 31: Whyte E, Romkes M, Mulsant BH, et al: CYP2D6 genotype and venlafaxine-xr concentrations in depressed elderly. Intl Geriatr Psychiatry 2006, in press 802