Paroxetine Associated Hepatotoxicity: A Report of 3 Cases and a Review of the Literature

• Case reports
• Case 1
• Case 2
• Case 3
• Discussion
• References

We recently encountered 3 patients who had developed reversible paroxetine-associated hepatotoxicity. Two of the patients were over 80 years old and their hepatitis was accompanied by hyponatremia. In the third case, hepatitis was associated with multiple organ failure and the co-administration of trazodone. Here, we will discuss the possible role of preexisting risk factors in the development of paroxetine hepatotoxicity and review the relevant literature.

Case reports

Case 1

A 49-year-old man suffering from major depressive disorder developed fever, vomiting and maculopapular rash 18 days after paroxetine (20 mg/day) and 8 days after trazodone (50 mg/day) were started. The patient was also given chloral hydrate (250 mg/day) for 22 days. Following the appearance of fever and rash, paroxetine, trazodone and chloral-hydrate were discontinued and fluvoxamine (50 mg/day) and mianserin (15 mg/day) were started. At that time, laboratory tests showed the following: AST 752 IU/l (normal: < 60 IU/l), ALT 975 (normal: 3 - 53 IU/l), Bilirubin 20 mcmol/l (normal: < 17 mcmol/l), GGTP 142 IU/l (normal: 10 - 80 IU/l), Na 133 mmol/l (normal: 135 - 145 mmol/l). Alkaline phosphatase, blood sodium, creatinine and urea were within normal range. Mild eosinophilia (7.9 %) and pneumonitis were also noticed. The next day, blood urea rose from 6.0 to 17.4 (normal: 3.3 - 6.5 mmol/l) and bilirubin from 20 to 59 mcmol/l. A skin biopsy revealed leukocytoclastic vasculitis. Mianserin and fluvoxamine were also discontinued. Hepatic and renal function continued to deteriorate. Nevertheless, liver ultrasound did not show any pathology. On day 3 after cessation of paroxetine and trazodone, the tests were as follows: AST 656 IU/l, ALT 3456 IU/l, Bilirubin 139 mcmol/l, GGTP 228 mmol/l, blood creatinine 549 mcmol/l (normal: 60 - 106 mcmol/l), urea 23.5 mmol/l (normal: 3.3 - 6.5 mmol/l), AlkP and blood sodium remained normal. On that day, high-dose steroid therapy was started along with fluids and diuretics due to the deterioration of the patient's condition. Nevertheless, cholestatic function tests continued to increase and reached their maximal levels on day 7 after discontinuation of paroxetine and trazodone and on day 4 after starting steroid treatment. On that day, alkaline phosphatase was 208 (normal: < 130 IU/l), bilirubin 148 mcmol/l and GGTP 810 IU/l. Serological tests for viral hepatitis A, B, C, CMV, EBV were negative and blood and urine cultures were sterile.

The patient's condition improved gradually. Skin rash and pneumonitis resolved themselves on days 4 and 7, respectively, whereas the renal function improved on day 10. Liver function tests returned to normal only 6 weeks later.

Case 2

An 80-year-old man was admitted to the Department of Internal Medicine presenting with coughing, bloody sputum and fever (38° C). The patient had been suffering from chronic ischemic heart disease and congestive heart failure for ten years. In addition, he complained about profound weakness of his left arm, recurrent falls and walking disturbances during the past 3 months. For several years, he had been receiving the following medications: aspirin 100 mg/day, isosorbide mononitrate at 20 mg × 2/day, captopril 12.5 mg × 3/day, verapamil 40 mg × 3/day, ranitidine 150 mg × 2/day and brotizolam 0.25 mg/day. Due to the patient's depressed mood, paroxetine 20 mg/day was added to this regimen six days before admission.

Laboratory tests on admission to our department were as follows: total serum protein was 51 g/l (normal: 60 - 80 g/l), albumin 23 g/l (normal: 35 - 50 g/l), and Na 124 mol/l (normal: 135 - 145 mol/l). Liver transaminases, alkaline phosphatase and bilirubin as well as complete blood count were normal. However, 15 days after the start of paroxetine therapy and 9 days after admission, ALT rose from 20 to 416 IU/l, AST from 21 to 693 IU/l, LDH from 628 to 3778 IU/l, urea from 4.1 mmol/l to 18.4 mmol/l, and creatinine from 55 mcmol/l to 161 mcmol/l. Liver ultrasound was normal, and physical examination did not reveal jaundice, ascites or epigastric tenderness. Paroxetine was stopped without changing any other medication. Five days later, AST and ALT returned to their normal range, and LDH decreased to 1482 IU/l. These values persisted during the rest of his stay in the hospital. In spite of the patient's improvement, CVA that occurred on day 11 after admission and pulmonary infection resistant to antibiotic treatment led to gradual deterioration of the patient's condition. Hyponatremia remained in spite of normal saline infusion and fluid restriction. His renal function deteriorated gradually, his creatinine levels rose from 61 on admission to 387 mcmol/l, urea from 5.4 to 26.9 mol/l, and on day 25 after admission, the patient died of sepsis.

Case 3

An 85-year-old man was admitted to the Department of Internal Medicine due to a confusional state and recurrent falls starting 12 hours prior to admission. Two months earlier, he had received irradiation therapy for an esophageal carcinoma. His medical history revealed chronic atrial fibrillation, congestive heart failure and hypertension. Over the last year, his medications had included digoxin at 0.125 mg × 1/day, warfarin according to INR 2.5, benazepril at 20 mg × 1/day, furosemide at 40 mg × 3/week. One week before hospitalization, he was put on paroxetine at 20 mg × 1/day by his family physician to treat his “bad mood”. On admission, his blood tests showed the following results: Na 123 mmol/l, K 3.3 mmol/l, glucose 9.0 mmol/l, hemoglobin 11.0, INR 2.7. His blood pressure was 135/70 mmHg. Benazepril and digoxin were stopped and normal saline was infused along with furosemide and drinking restriction. The patient had brought paroxetine with him and continued to take his regular dose for additional 6 days without the attention of his treating physicians. On the 6th day, the first slight elevation of liver enzymes was noted. ALT was 61 IU/l. AST was 101 IU/l, LDH 936 IU/l, Bilirubin 26 mcmol/l. Paroxetine was stopped, but the increase in liver enzymes continued, peaking on the next day: ALT was 630 IU/l, AST 1566 IU/l, LDH 3786 IU/l. Alkaline phosphatase (AlkP) and bilirubin reached their maximal levels after two days, 152 IU/l and 45 mcmol/l, respectively. However, liver ultrasound remained within normal limits. The day after discontinuation of paroxetine, deterioration of renal function was observed (creatinine rose from 63 mcmol/l on admission to 151 mcmol/l, urea from 3.5 mmol/l to 15.4 mmol/l). However, one week later, the renal functions and the serum sodium level returned to normal. Liver function tests returned to normal two weeks later. Digoxin, warfarin and furosemide were reintroduced, and lisinopril and famotidine was started. Liver function test results remained within normal range during the next 4 weeks.

Discussion

Paroxetine belongs to the family of selective serotonin reuptake inhibitors (SSRIs). Since early 90's, it has been used for the treatment of major depressive disorder, social phobia, panic and obsessive-compulsive disorders [23]. Because of its favorable safety profile and relatively short half life (12 - 24 hr), it is one of the drugs of choice for the treatment of the conditions mentioned in the elderly [12] [23]. The most commonly reported side effects include gastrointestinal disturbances such as nausea, vomiting, flatulence, diarrhea and epigastric discomfort, CNS disturbances such as headaches, dizziness, paraesthesia, tremor, anxiety, agitation and confusion, as well as skin eruptions and the syndrome of inappropriate antidiuretic hormone secretion (SIADH). The frequency of those side effects is dose-related [3] [12].

According to a computer-assisted search of the literature (MEDLINE 1966 - 2000) and the data on file at the Drug Information Center of Clinical Pharmacology Department at the Hadassah-Hebrew University Medical Center, hepatotoxicity is a relatively rare side effect of paroxetine. Transient mild elevations of liver enzymes have been observed since the early clinical experience with paroxetine [9] [15] [22]. Usually, transaminases were normalized few days following cessation of the drug. The Committee on Safety of Medicines has been notified of 54 cases of abnormal levels of liver enzymes without evidence of hepatitis [2]. However, till now only 5 cases of paroxetine-induced hepatitis have been published in the English medical literature [2] [6] [11] [18].

Here we present 3 new cases of hepatotoxicity associated with the use of paroxetine. In all of them, other common causes of liver injury have been excluded by negative serological tests for hepatitis A, B, C and HSV, EBV, CMV as well as negative microbiological tests on blood and urine. The patients were not alcoholic, and their liver ultrasound showed no gross pathology. Liver biopsy has not been done in any of the cases. The relatively quick improvement of liver function tests (LFT) after cessation of paroxetine strongly suggested its offending role.

In the first case, hepatotoxicity developed 18 days after starting paroxetine and was accompanied by leukocytoclastic vasculitis, pneumonitis and eosinophilia. The concomitant drug that also could attribute to the condition was trazodone. This is a non-tricyclic antidepressant drug and has been marketed in USA since 1983. It possesses mixed serotoninergic, alpha-blocking and antihistamine activity. Few reports of hepatotoxic reactions include intrahepatic cholestasis [7] and mixed hepatocellular damage of mild severity [4] that started within 10 - 20 days of trazodone therapy. A case of chronic active hepatitis after 8 month of trazodone therapy [1] and acute hepatitis with jaundice, liver enzyme elevation, pruritus, nausea and leukonychia after a 7-month course were reported [17]. Leukocytoclastic vasculitis after 3 months of trazodone use has also been described [19]. Since our patient received trazodone 8 days before the appearance of toxic reactions, its role in this complication cannot be excluded. Nevertheless, the concomitant appearance of hyponatremia and disturbed liver function tests favor the possibility of paroxetine-induced hepatitis. In the second case, the elevation of transaminases was not accompanied by cholestatic disturbances. The LFT returned back to normal following the discontinuation of paroxetine despite adding antibiotics and otherwise keeping to the medication regime. This observation suggests that the offending drug in this case was paroxetine.

In the third case, alkaline phosphatase and bilirubin were slightly elevated (153 IU/l and 45 mcmol/l correspondingly), but there was no evidence of jaundice, and the LFT returned to normal following the discontinuation of paroxetine. Readministration of famotidine and benazepril did not result in hepatitis, which again suggests that paroxetine was the offending drug in the third case as well.

The characterizing features of our 3 cases and the other 5 cases from the literature are summarized in Table [1]. In summary, there were 4 women and 4 men with ages ranging from 31 years to 85 years. The time between the start of the paroxetine therapy and appearance of hepatotoxic symptoms ranged from 15 days to 17 months. In three cases presented by us, concomitant hyponatremia and nephrotoxicity 24 h after cessation of paroxetine therapy were observed. No other causes of disturbed renal function were found. While hyponatremia due to SSRIs is the thoroughly documented reversible side effect of paroxetine with a known mechanism [20], nephrotoxicity in association with this medicine could hardly be found [14] [21]. Based upon our current knowledge of its renal elimination properties, we cannot explain this observation. As skin vasculitis and pneumonitis were found in case 1, kidney involvement may also be part of a systemic vasculitic disease. However, this explanation cannot stand for the remaining two cases.

Less than 2 % of an oral dose of paroxetine is excreted unchanged in urine and feces, but 98 % is excreted as metabolites (conjugates and other unknown polar metabolites). Approximately 65 % of a dose is excreted in the urine and the rest (35 - 37 %) in feces [14]. In a single-dose study involving subjects with varying degrees of renal impairment, maximum serum levels and area under the concentration-time curve (AUC) values of paroxetine tended to increase as renal function declined. The elimination half-life was prolonged significantly only in severe renal impairment (creatinine clearance less than 30 milliliters/minute) [8].

All abnormalities observed in these cases associated with the use of paroxetine resolved following the cessation of the drug. The resolution time ranged from one week to 6 months and showed a good correlation with the period of paroxetine administration (total cumulative dose), preexisting disease (cirrhosis, case 4) and concomitant use of other hepatotoxic drugs like trazodone (case 1) and atrium (cases 6 and 7, atrium is a combination of phenobarbital, febarbamate and difebarbamate). Drug-induced hepatitis is generally unpredictable. Hepatotoxicity may be related to the drug itself, or to chemically reactive metabolites which can bind covalently to hepatic macromolecules and may lead to either idiosyncratic, toxic hepatitis or to immunoallergic hepatitis [10]. Drug hepatotoxicity is partially determined by genetic factors involved in drug metabolism that may, for example, involve the debrisoquine oxidation polymorphism mediated by cytochrome (CYP) 2D6 [16]. Metabolism of paroxetine depends on the genetic polymorphism of CYP2D6, and the drug itself is also a potent inhibitor of this enzyme [21]. There is now evidence indicating that genetic variations in systems of biotransformation or detoxication may modulate either the toxic or sensitizing effects of some drugs. Recent studies have shown that genetic deficiency in a particular hepatic cytochrome P 450 isozyme (CYP 2D6) is involved in perhexiline liver injury [10] [20]. The deficiency in CYP 2C19 and CYP2CMP might also contribute to atrium hepatotoxicity [13]. At the end, the few lessons that we learned from these cases should be emphasized. It may be advisable to determine GLDH (glutamate dehydrogenase) blood levels as an additional marker for drug-induced hepatotoxicity [5]. Second, paroxetine in two of the above cases was continued despite the finding of hyponatremia and a confusional state in one of them. Since this drug can cause SIADH, it is important to stop this medication when hyponatremia is detected. The third indication is to avoid starting alternative medications while adverse effects are still in progress (as happened in case 1). This is because we can not appreciate the exact contribution of the new drug to the progression of these side effects.

In summary, the cases of paroxetine hepatotoxicity presented above would suggest an allergic or idiosyncratic drug reaction that is unpredictable and usually does not require LFT monitoring. However, the appearance of more severe forms of paroxetine-associated hepatic damage in the presence of other risk factors such as preexisting cirrhosis (case 4), other hepatotoxic drugs (trazodone, case no 1, atrium - cases 6 and 7), or total cumulative dose do warrant careful monitoring of LFT in such cases. Paroxetine-attributed hepatotoxicity observed in the absence of these factors is relatively mild and more rapidly reversible (1 - 2 weeks vs. 3 - 6 months), even in critically ill patients. Clinical awareness and immediate dechallenge of the drug are important for rapid recovery.

References

• 1 Beck P L, Bridges R J, Demetrick  D J. et al . Chronic active hepatitis associated with trazodone therapy.  Ann Intern Med. 1993;  118 10-16
  PubMedSearch in Google Scholar
• 2 Benbow S J, Gill G. Paroxetine and hepatotoxicity.  BMJ. 1997;  314 s1387
  PubMedSearch in Google Scholar
• 3 Boyer W F, Blumhardt C L. The safety profile of paroxetine.  J Clin Psychiatry. 1992;  53 (6) 61-66
  PubMedSearch in Google Scholar
• 4 Brogden R N, Heel R C. et al . Trazodone: review pharmacological properties and therapeutic use in depression and anxiety Drugs.  . 1981;  21 401-429
  PubMedSearch in Google Scholar
• 5 Brondeau M T, Ban M, Bonnet P, Guenier J, De Ceaurriz J. Concentration-related changes in blood and tissue parameters of hepatotoxicity and their interdependence in rats exposed to bromobenzene and 1,2-dichlorobenzene.  Toxicol Lett. 1986;  31 (2) 159-166
  CrossrefPubMedSearch in Google Scholar
• 6 Cadranel J F, Martino V, Cozier A. et al . Atrium and paroxetine-related severe hepatitis.  J Clin Gastroenterol. 1999;  28 52-55
  CrossrefPubMedSearch in Google Scholar
• 7 Chu A G, Gansolly B L, Summers R W. et al . Chronic active hepatitis associated with trazodone therapy.  Ann Intern Med. 1983;  99 128-129
  PubMedSearch in Google Scholar
• 8 Doyle G D, Laher M, Kelly J G. et al . The pharmacokinetics of paroxetine in renal impairment.  Acta Psychiatr Scand. 1989;  80 (350) 89-90
  PubMedSearch in Google Scholar
• 9 Dunbar G C, Claghorn J L, Kiev A. et al . A comparison of paroxetine and placebo in depressed patients.  Acta Psychiatr Scand. 1993;  87 302-305
  PubMedSearch in Google Scholar
• 10 Fenichel R R. Drug-induced hepatotoxicity.  N Engl J Med. 1996;  334 (13) 864
  CrossrefPubMedSearch in Google Scholar
• 11 Helmchen C, Boerner R J, Meyendorf R. et al . Reversible Hepatotoxicity of paroxetine in a patient with major depression.  Pharmacopsychiatry. 1996;  29 223-226
  PubMedSearch in Google Scholar
• 12 Hiemke C, Hartler S. Pharmacokinetics of selective serotonin reuptake inhibitors.  Pharmacol Ther. 2000;  85 (1) 11-28
  CrossrefPubMedSearch in Google Scholar
• 13 Horsmans Y, Lannes D, Pessayre D, Larrey D. Possible association between poor metabolism of mephenytoin and hepatotoxicity caused by Atrium, a fixed combination preparation containing phenobarbital, febarbamate and difebarbamate.  J Hepatol. 1994;  21 (6) 1075-1079
  CrossrefPubMedSearch in Google Scholar
• 14 Kaye C M, Haddock R E, Langley P F. et al . A review of the metabolism and pharmacokinetics of paroxetine in man.  Acta Psychiatr Scand. 1989;  80 (350) 60-75
  PubMedSearch in Google Scholar
• 15 Kuhs H, Rudolf G AE. A double-blind study of the comparative antidepressant effect of paroxetine and amitriptyline.  Acta Psychiatr Scand. 1989;  80 (350) 89-90
  PubMedSearch in Google Scholar
• 16 Larrey D, Pageaux G P. Genetic predisposition to drug-induced hepatotoxicity.  J Hepatol. 1997;  26 (2) 12-21
  CrossrefPubMedSearch in Google Scholar
• 17 Longstreth G F, Hershman J. Trazodone-induced hepatotoxicity and leukonychia.  J Amer Acad Dermatol. 1985;  13 149-150
  PubMedSearch in Google Scholar
• 18 Man R A. Severe hepatitis attributed to paroxetine. In Dutch.  Ner Tijdschr Geneeskd. 1997;  141 (11) 540-542
  PubMedSearch in Google Scholar
• 19 Mann S C, Walker M M, Messenger G G, Greenstein R A. Leukocytoclastic vasculitis secondary to trazodone treatment.  J Am Acad Dermatol. 1984;  10 (4) 669-670
  PubMedSearch in Google Scholar
• 20 Meynaar I A, Peelers A J, Mulder A H. et al . Syndrome of inappropriate ADH secretion attributed to the serotonin re-uptake inhibitors, venlafaxine and paroxetine.  Neth J Med. 1997;  50 243-245
  PubMedSearch in Google Scholar
• 21 Rasmussen B B, Brossen K. Is therpeutic drug monitoring a case for optimizing clinical outcome and avoiding interactions of the selective serotonin reuptake inhibitors.  Ther Drug Monit. 2000;  22 (2) 143-154
  CrossrefPubMedSearch in Google Scholar
• 22 Rickels K, Amsterdam J, Clary C. et al . A placebo-controlled, double blind clinical trial of paroxetine in depressed outpatients.  Acta Psychiatr Scand. 1989;  80 (350) 117-123
  PubMedSearch in Google Scholar
• 23 Schatzberg A F. New indications for antidepressants.  J Clin Psychiatry. 2000;  61 (11) 9-17
  PubMedSearch in Google Scholar

Tanya Azaz-LivshitsPhD 

Dept. of Medicine

PO Box 12000

Jerusalem 91120

Israel

Email: azaz@hadassah.org.il

References

• 1 Beck P L, Bridges R J, Demetrick  D J. et al . Chronic active hepatitis associated with trazodone therapy.  Ann Intern Med. 1993;  118 10-16
  PubMedSearch in Google Scholar
• 2 Benbow S J, Gill G. Paroxetine and hepatotoxicity.  BMJ. 1997;  314 s1387
  PubMedSearch in Google Scholar
• 3 Boyer W F, Blumhardt C L. The safety profile of paroxetine.  J Clin Psychiatry. 1992;  53 (6) 61-66
  PubMedSearch in Google Scholar
• 4 Brogden R N, Heel R C. et al . Trazodone: review pharmacological properties and therapeutic use in depression and anxiety Drugs.  . 1981;  21 401-429
  PubMedSearch in Google Scholar
• 5 Brondeau M T, Ban M, Bonnet P, Guenier J, De Ceaurriz J. Concentration-related changes in blood and tissue parameters of hepatotoxicity and their interdependence in rats exposed to bromobenzene and 1,2-dichlorobenzene.  Toxicol Lett. 1986;  31 (2) 159-166
  CrossrefPubMedSearch in Google Scholar
• 6 Cadranel J F, Martino V, Cozier A. et al . Atrium and paroxetine-related severe hepatitis.  J Clin Gastroenterol. 1999;  28 52-55
  CrossrefPubMedSearch in Google Scholar
• 7 Chu A G, Gansolly B L, Summers R W. et al . Chronic active hepatitis associated with trazodone therapy.  Ann Intern Med. 1983;  99 128-129
  PubMedSearch in Google Scholar
• 8 Doyle G D, Laher M, Kelly J G. et al . The pharmacokinetics of paroxetine in renal impairment.  Acta Psychiatr Scand. 1989;  80 (350) 89-90
  PubMedSearch in Google Scholar
• 9 Dunbar G C, Claghorn J L, Kiev A. et al . A comparison of paroxetine and placebo in depressed patients.  Acta Psychiatr Scand. 1993;  87 302-305
  PubMedSearch in Google Scholar
• 10 Fenichel R R. Drug-induced hepatotoxicity.  N Engl J Med. 1996;  334 (13) 864
  CrossrefPubMedSearch in Google Scholar
• 11 Helmchen C, Boerner R J, Meyendorf R. et al . Reversible Hepatotoxicity of paroxetine in a patient with major depression.  Pharmacopsychiatry. 1996;  29 223-226
  PubMedSearch in Google Scholar
• 12 Hiemke C, Hartler S. Pharmacokinetics of selective serotonin reuptake inhibitors.  Pharmacol Ther. 2000;  85 (1) 11-28
  CrossrefPubMedSearch in Google Scholar
• 13 Horsmans Y, Lannes D, Pessayre D, Larrey D. Possible association between poor metabolism of mephenytoin and hepatotoxicity caused by Atrium, a fixed combination preparation containing phenobarbital, febarbamate and difebarbamate.  J Hepatol. 1994;  21 (6) 1075-1079
  CrossrefPubMedSearch in Google Scholar
• 14 Kaye C M, Haddock R E, Langley P F. et al . A review of the metabolism and pharmacokinetics of paroxetine in man.  Acta Psychiatr Scand. 1989;  80 (350) 60-75
  PubMedSearch in Google Scholar
• 15 Kuhs H, Rudolf G AE. A double-blind study of the comparative antidepressant effect of paroxetine and amitriptyline.  Acta Psychiatr Scand. 1989;  80 (350) 89-90
  PubMedSearch in Google Scholar
• 16 Larrey D, Pageaux G P. Genetic predisposition to drug-induced hepatotoxicity.  J Hepatol. 1997;  26 (2) 12-21
  CrossrefPubMedSearch in Google Scholar
• 17 Longstreth G F, Hershman J. Trazodone-induced hepatotoxicity and leukonychia.  J Amer Acad Dermatol. 1985;  13 149-150
  PubMedSearch in Google Scholar
• 18 Man R A. Severe hepatitis attributed to paroxetine. In Dutch.  Ner Tijdschr Geneeskd. 1997;  141 (11) 540-542
  PubMedSearch in Google Scholar
• 19 Mann S C, Walker M M, Messenger G G, Greenstein R A. Leukocytoclastic vasculitis secondary to trazodone treatment.  J Am Acad Dermatol. 1984;  10 (4) 669-670
  PubMedSearch in Google Scholar
• 20 Meynaar I A, Peelers A J, Mulder A H. et al . Syndrome of inappropriate ADH secretion attributed to the serotonin re-uptake inhibitors, venlafaxine and paroxetine.  Neth J Med. 1997;  50 243-245
  PubMedSearch in Google Scholar
• 21 Rasmussen B B, Brossen K. Is therpeutic drug monitoring a case for optimizing clinical outcome and avoiding interactions of the selective serotonin reuptake inhibitors.  Ther Drug Monit. 2000;  22 (2) 143-154
  CrossrefPubMedSearch in Google Scholar
• 22 Rickels K, Amsterdam J, Clary C. et al . A placebo-controlled, double blind clinical trial of paroxetine in depressed outpatients.  Acta Psychiatr Scand. 1989;  80 (350) 117-123
  PubMedSearch in Google Scholar
• 23 Schatzberg A F. New indications for antidepressants.  J Clin Psychiatry. 2000;  61 (11) 9-17
  PubMedSearch in Google Scholar

Tanya Azaz-LivshitsPhD 

Dept. of Medicine

PO Box 12000

Jerusalem 91120

Israel

Email: azaz@hadassah.org.il