- Email: [email protected]
Valproate-Induced Hyperammonemic Encephalopathy in General Hospital Patients With One or More Psychiatric Disorders
Author’s Accepted Manuscript Valproate-induced Hyperammonemic Encephalopathy (VHE) in General Hospital Patients with one or more Psychiatric DisordersValproate-induced hyperammonemic encephalopathy in a general hospital☆ Chandani Lewis, George E. Tesar, Roman Dale www.elsevier.com/locate/psym
PII: DOI: Reference:
S0033-3182(17)30041-5 http://dx.doi.org/10.1016/j.psym.2017.02.003 PSYM738
To appear in: Psychosomatics Cite this article as: Chandani Lewis, George E. Tesar and Roman Dale, Valproate-induced Hyperammonemic Encephalopathy (VHE) in General Hospital Patients with one or more Psychiatric DisordersValproate-induced hyperammonemic encephalopathy in a general hospital☆, Psychosomatics, http://dx.doi.org/10.1016/j.psym.2017.02.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ORIGINAL ARTICLE Valproate-induced hyperammonemic encephalopathy (VHE) in general hospital patients with one or more psychiatric disorders Chandani Lewis 1,2, George E. Tesar 1, Roman Dale 1 1
Dept. of Psychiatry and Psychology, Neurological Institute, Cleveland Clinic, Cleveland, OH 44195
USA 2
Current address: Medical Department, Zepf Center, 6605 W Central Ave, Toledo, OH 43615 & Dept.
of Psychiatry, University of Toledo College of Medicine and Life Sciences, Toledo, OH 43615, USA.
Address for Correspondence: Chandani Lewis, MD Medical Department Zepf Center, 6605 W Central Ave Toledo, OH 43615 419-383-5683 Phone: (419) 383-5683; Fax: (419) 383- 5630; Email: [email protected]
Short Title: Valproate-induced hyperammonemic encephalopathy in a general hospital
Conflicts of interest: Nothing to declare for all authors. Date:
1
Financial Disclosure No direct funding was received for this study. No specific salary was set aside or given for the writing of this paper. No funding bodies had any role in the study design, data collection, analysis, decision to publish or preparation of the manuscript.
Abstract Background: Divalproex sodium/valproic acid (VPA) is an antiepileptic drug approved for use in epilepsy and bipolar disorder. Valproate-induced hyperammonemia (VHA) occurs in up to 50 % of VPA-treated patients, some of which may become encephalopathic. Valproate-induced encephalopathy (VHE) is thought to be rare, and for a variety of reasons the diagnosis requires a high index of suspicion. The study’s goals are to determine how common VHE is, and the quality of treatment provided when diagnosed.
Methods: Retrospective, cross-sectional survey of general hospital patients. The hospital’s laboratory and pharmacy databases were combined to identify a cohort of all VPA-treated patients during a five-year period that developed hyperammonemia. Hospital records of the subset of patients with a psychiatric disorder were selected and reviewed for data collection. Results: Twenty of 793 VPA-treated patients (2.52%) had signs and symptoms consistent with VHE.
The majority were Caucasian males on multiple psychotropic agents. Valproate was
appropriately discontinued in eight (40%). Lactulose was the only ammonia-lowering drug used, and administered to six patients only one of who also had VPA discontinued.
2
Conclusion: Study results indicate that valproate-induced hyperammonemic encephalopathy (VHE) may be more common in psychiatric patients than previously assumed, but under-recognized and inadequately treated. The diagnosis of VHE requires a high-index of suspicion.
Outcome is
favorable once it is recognized and treated appropriately.
Introduction Valproate-induced hyperammonemic encephalopathy (VHE), recently reviewed by our group and another (1, 2), is recognized as a rare, but serious and unpredictable adverse outcome of treatment with valproic acid (VPA) including its congener, divalproex sodium. Most of the existing VHE literature is based on treatment of epilepsy (3, 4); there are fewer reports of its use for bipolar disorder or off-label indications such as anxiety, agitation, or delirium (2, 5). Clinical findings of VHE are nonspecific, and typically present acutely or subacutely as drowsiness, lethargy, agitation, confusion or coma. Other features include ataxia, vomiting, focal neurological signs, low grade fever, and seizures (1, 2, 5).
Recognition of VHE requires a high level of clinical suspicion, since clinical presentation is nonspecific and correlates poorly with dosage, blood levels or duration of VPA treatment (5). The transient, mild liver enzyme elevation often seen during VPA therapy is are not necessarily present, and, similar to hepatic encephalopathy, correlation of VHE severity and blood ammonia levels is variable (5). Asymptomatic ammonia elevations are seen in as many as 51% of VPA-treated psychiatric patients (6); in contrast, normal serum ammonia levels have been reported in unusual cases of VHE (5).
3
Our impression is that VHE may be more common than previously reported. To test this assumption we investigated a sample of general hospital psychiatric patients. In addition to frequency of occurrence of VHE, we were interested in its rate of recognition, and the quality of treatment provided when recognized. To our knowledge this is the first study to examine VHE frequency in a discrete patient population.
Methods This is a retrospective, case-based, cross-sectional survey of general hospital patients treated with divalproex sodium or valproic acid (VPA) during a five-year period, 2004 to 2009. The cohort of interest was patients with one or more psychiatric disorders who developed VHE. The list of study patients was generated by comparing the hospital’s pharmacy and laboratory databases to identify all VPA-treated patients who also had evidence of hyperammonemia during their index admission. In order to reduce the likelihood of false positives, hyperammonemia was defined as a blood ammonia level of 41 µmol/L or more (that is, 15% above the upper limit of normal, 35 µmol/L). Hyperammonemic patients with at least one primary psychiatric diagnosis were selected for detailed chart review.
Two investigators (CL and RD) reviewed clinical records for documented evidence of an acute change in mental status consistent with a diagnosis of encephalopathy/delirium. Key search terms included agitation, altered sensorium or mental status, cognitive deficits, confusion, lethargy, delirium, delusions, dementia, disorientation, falls, hallucinations, memory impairment, and speech abnormalities.
4
Patients with hepatic enzyme elevations (aspartate transaminase [AST] and alanine aminotransferase [ALT]) were excluded to eliminate hepatic failure as a cause of hyperammonemia.
This study was approved by the Cleveland Clinic Institutional Review Board.
Results
During the five-year study period the hospital’s pharmacy database listed 793 VPA-treated inpatients; the laboratory database included 294 patients with one or more instances of hyperammonemia (blood ammonia level > 41 µmol/L). Pooling these data identified 30 VPA-treated patients with at least one report of hyperammonemia. Ten of these patients were excluded from further study for a variety of reasons, including elevated AST and ALT blood levels (n=7), seizure disorder alone with no associated psychiatric diagnosis (n=2), and insufficient evidence of encephalopathy (n=1). Detailed record review was performed in the remaining 20.
Demographic, diagnostic, treatment, and laboratory data are listed in the Table. The average age of the patients was 59.55 years+19.35 (range, 21-89). Fifteen (75%) were male, 14 were Caucasian (70%), and six were African American (30%). Twelve (60%) were admitted to the general medicine service and eight were admitted to either an adult (n=4) or geriatric inpatient psychiatry unit (n=4).
The 20 patients collectively had 31 psychiatric diagnoses including schizophrenia or schizoaffective disorder (n=9), dementia (n=6), major depressive disorder (n=5), bipolar disorder (n=4), current or a history of substance abuse disorder (n=4), intellectual disability (n=3), and intermittent explosive
5
disorder (1). Seven had a single psychiatric diagnosis, including two who also had a diagnosis of intellectual disability.
The most common medical co-morbidities were hypertension (n=11), diabetes mellitus (n=5), epilepsy (n=5), cardiovascular or peripheral vascular diseases (4), and chronic kidney disease (3).
The average number of psychoactive drugs per patient was 2.8+1.39 (range 1-6). In addition to VPA, patients received other mood stabilizers: oxcarbazepine (n=3), lithium (n=3), lamotrigine (n=2), and topiramate (n=1); antipsychotics: quetiapine (n=7), risperidone (n=3), aripiprazole (n=2), ziprasidone (n=2), olanzapine, haloperidol and clozapine, (n=1 each); antidepressants: serotonin selective reuptake inhibitors (n=8) and mirtazapine (n=2); or anxiolytics: benzodiazepines (n=6), and gabapentin (n=3).
The mean daily VPA dose was 1280 mg+696.5 (range, 250-3000 mg per day). Blood levels of VPA were available in 10 patients with a mean level of 81.7µg/l+ 6.79 (range, 18 –289µg/l). All VPA blood levels were within therapeutic range except in the patient who overdosed (289µg/l).
Ammonia levels were drawn in the morning in all but three patients with blood levels ranging from 43205 (median, 63); only two patients – one of who overdosed on VPA – had serum ammonia levels > 100 µmol).
Hyperammonemia was first detected after a variable period following initiation of VPA treatment: within the first month (n=4); within one year (n=1); between one and five years (n=9); between six to 10 years (n=3); and for an unknown duration (n=3).
6
In addition to typical signs and symptoms of encephalopathy, two patients were diagnosed with presumptive treatment-resistant seizures.
Despite evidence of hyperammonemia and encephalopathy 12 patients (60%) continued to receive VPA, including one whose dosage was increased. Lactulose was administered to six patients (30%) only one of who also stopped VPA. No patient received alternative agents indicated for treatment of hyperammonemia (e.g., levo-carnitine, charcoal, neomycin, or rifamixin). .
Discussion
Our study suggests that the occurrence of VHE in psychiatric patients is by no means rare, and yet often overlooked. In the population studied, 20 of 793 VPA-treated patients (2.52%) developed retrospectively diagnosed VHE. This may in fact underestimate the true frequency of VHE given that hyperammonemia rather than a diagnosis of delirium or encephalopathy was the main selection criterion for study participation. This methodological limitation over-looks VPA-treated patients who may have developed VHE but had either a normal ammonia level, an ammonia level between 36 to 40 µmol, or did not have blood ammonia checked at all. Gerstner and colleague’s meta-analysis of 51 cases of VHE reported during a 10 year period in patients with epilepsy (7) may provide a point of comparison. Assuming a rate of VPA-treatment similar to our group the average number of cases per year identified by Gerstner et al, which is, five, is nearly the same as the four per year identified in our study .
7
Although the study’s methodology prevents firm conclusions regarding correlations between VHE and a variety of variables, including age, gender, ammonia levels and duration or intensity of VPA treatment, our results are largely consistent with previous reports.
In this exclusively adult population, VHE occurred across the life-span (ages 21- 89) with seven patients (35%) aged 65 or older. Since other age-related variables increase the likelihood of preexisting cognitive impairment, VPA-treated patients older than 65 should have blood ammonia levels checked both before and after starting VPA treatment. Accurate interpretation of ammonia blood levels in the elderly, however, should be informed by a recent study reporting a two-fold increase in serum ammonia levels in older psychiatric patients (8).
The duration of VPA therapy varied from a few days to 10 years. Among the 10 patients with documented VPA blood levels, all were within normal range except for one from the patient who overdosed on VPA. The apparent lack of correlation between VPA blood levels and hyperammonemia or the likelihood of developing VHE supports previous reports (3-5, 9)
Only two of our patients (10%) had ammonia levels exceeding 100 µmols/L, which parallels previous studies showing poor correlation between ammonia level and VHE presence or severity. Both Dealberto et al (5) and Verrotti et al (4) reported a positive correlation between duration of VPA therapy and serum ammonia levels, but, as noted, ammonia level is a poor predictor of clinical presentation. .
8
Patients in our study were on an average of three psychoactive drugs with a maximum number of six. Polypharmacy (> 3 agents) has been identified as a risk factor for developing hyperammonemia (2, 5) and may increase the risk of VHE by interfering with normal urea cycle activity, reducing carnitine levels, or displacing highly protein-bound VPA (1). Simultaneous use of VPA and other drugs reported to cause VHE (e.g., topiramate, lamotrigine) or pharmacokinetic interaction with VHA (e.g., risperidone) should be avoided or monitored carefully (10-13).
Detection and management of VHE in our population was suboptimal similar to the experiences reported by others (2, 5). Diagnosis of VHE warrants either discontinuation of VPA or dosage reduction depending on symptom severity (6). Treatment with VPA was discontinued in only eight of our patients (40%) and the dosage was increased in one patient underscoring the importance of maintaining a high index of suspicion for VHE.
Presenting as either hypoactive or hyperactive
delirium, VHE can be confused with depression or manic agitation potentially contributing to the misinterpretation of inadequate VPA-treatment (6, 14). While severe VHE can present with epileptic seizures (4) it is more likely to present with non-epileptic events either misdiagnosed as epileptic or as an exacerbation of a primary psychiatric disorder. Any of these circumstances can lead to inappropriate escalation rather than de-escalation of VPA-treatment. For example, two of our patients (case 4 & 5) with “uncontrolled seizures” continued to receive VPA when it should have been discontinued.
Persistence of VHE despite reduction or discontinuation of VPA is an indication for additional ammonia–depleting agents such as lactulose, charcoal, neomycin, rifamixin, or levo-carnitine (6, 15). Five patients (30%) received any kind of treatment, which in all instances was lactulose. None of our patients required hemodialysis (indicated for ammonia blood levels >400 µmol/L) (16). There were no
9
reported deaths and all of our patients were discharged from the hospital similar to findings in other reports (5, 17).
Interpretation of our findings must be considered with the following study limitations in mind. The cross-sectional, small sample size, retrospective design, as previously noted, prevented firm conclusions about predictors of VHE. Additionally, the lack of a gold-standard diagnostic instrument or a clinical finding pathognomonic of VHE compromises diagnostic validity thereby increasing the likelihood of either false-positive or false-negative diagnoses. As discussed previously, however, the likelihood of false-negatives (under-diagnosis) is probably greater than false-positives (over-diagnosis) supporting the contention that true VHE frequency is likely greater than the 2.52% identified in our population. Availability of EEG recordings may have helped improve diagnostic validity, but it was unlikely to facilitate differentiation of VPA from other causes of encephalopathy.
Systematic
assessment of ammonia levels in all VPA-treated patients including those who did not develop VHE prevented firm conclusions about the relationship between VHE and blood ammonia levels.
Conclusion: During a five-year study interval valproate-induced hyperammonemic encephalopathy (VHE) was identified in 2.52% of valproate (VPA)-treated general hospital psychiatric patients. Study methodology very likely contributed to under-estimation of the true frequency of VHE and only a minority of patients with VHE received adequate treatment. Our findings indicate a need for increased familiarity with the diagnosis and with the appropriate treatment of VHE.
10
References: 1. Lewis C, Deshpande A, Tesar GE, Dale R. Valproate-induced hyperammonemic encephalopathy: a brief review. Curr Med Res Opin. 2012;28(6):1039-42. 2. Chopra A, Kolla BP, Mansukhani MP, Netzel P, Frye MA. Valproate-induced hyperammonemic encephalopathy: an update on risk factors, clinical correlates and management. Gen Hosp Psychiatry. 2012;34(3):290-8. 3. Segura-Bruna N, Rodriguez-Campello A, Puente V, Roquer J. Valproate-induced hyperammonemic encephalopathy. Acta Neurol Scand. 2006;114(1):1-7. 4. Verrotti A, Trotta D, Morgese G, Chiarelli F. Valproate-induced hyperammonemic encephalopathy. Metab Brain Dis. 2002;17(4):367-73. 5. Dealberto MJ. Valproate-induced hyperammonaemic encephalopathy: review of 14 cases in the psychiatric setting. Int Clin Psychopharmacol. 2007;22(6):330-7. 6. Raja M, Azzoni A. Valproate-induced hyperammonaemia. J Clin Psychopharmac. 2002;22(6):631-3. 7. Gerstner T, Buesing D, Longin E, Bendl C, Wenzel D, Scheid B, et al. Valproic acid induced encephalopathy--19 new cases in Germany from 1994 to 2003--a side effect associated to VPA-therapy not only in young children. Seizure. 2006;15(6):443-8. 8. Adler LW, Regenold WT. Valproate-Related Hyperammonemia in Older Adult Psychiatric Inpatients. Prim Care Companion CNS Disord. 2015;17(2). 9. Mehndiratta MM, Mehndiratta P, Phul P, Garg S. Valproate induced non hepatic hyperammonaemic encephalopathy (VNHE)--a study from tertiary care referral university hospital, north India. JPMA J Pak Med Assoc. 2008;58(11):627-31. 10. Latour P, Biraben A, Polard E, Bentue-Ferrer D, Beauplet A, Tribut O, et al. Drug induced encephalopathy in six epileptic patients: topiramate? valproate? or both? Hum Psychopharmacol. 2004;19(3):193-203. 11. Kimmel RJ, Irwin SA, Meyer JM. Valproic acid-associated hyperammonemic encephalopathy: a case report from the psychiatric setting. Int Clin Psychopharmacol. 2005;20(1):57-8. 12. Reif A, Leonhard C, Mossner R, Lesch KP, Fallgatter AJ. Encephalopathy and myoclonus triggered by valproic acid. Prog Neuropsychopharmacol Biol Psychiatry. 2004;28(6):1061-3. 13. Carlson T, Reynolds CA, Caplan R. Case report: valproic Acid and risperidone treatment leading to development of hyperammonemia and mania. J Am Acad Child Adolesc Psychiatry. 2007;46(3):356-61. 14. Carr RB, Shrewsbury K. Hyperammonemia due to valproic acid in the psychiatric setting. Am J Psychiatry. 2007;164(7):1020-7. 15. Lheureux PE, Hantson P. Carnitine in the treatment of valproic acid-induced toxicity. Clin. Toxicol. 2009;47(2):101-11. 16. Eyer F, Felgenhauer N, Gempel K, Steimer W, Gerbitz KD, Zilker T. Acute valproate poisoning: pharmacokinetics, alteration in fatty acid metabolism, and changes during therapy. J Clin Psychopharmacol. 2005;25(4):376-80. 17. Raby WN. Carnitine for valproic acid-induced hyperammonemia. Am J Psychiatry. 1997;154(8):1168-9.
11
TABLE. Demographic, diagnostic, laboratory measures and treatment in 20 patients with VHE C as e
Age/ Sex1
Ra ce2
Psychiatric diagnosis
Medical Co-morbidities
VPA dose (max mg/d)
VPA blood level (µg/l)
Ammonia serum level (µmol)
VPA duration (yr/mo)
VHE Treatmen t3 Lac VP tuA4 lose 5
1
89/F
C
MDD, Alzheimer’s Dementia
2
44/ M
C
BPD
DM, HTN
500
C
Paranoid schizophrenia
DM, HTN
1000
C
BPD
DM, HTN, Epilepsy
2000
3 4
60/ M 52/ M
5
34/ M
C
MDD, SA, personality disorder
6
64/ M
C
Schizophrenia, ID
7
44/ M
C
Schizoaffective
8
89/ M
C
9
67/ M
C
1 0
87/ M
C
1 1
64/F
1 2 1 3 1 4 1 5 1 6
Alzheimer’s dementia, psychosis Dementia, h/o alcohol abuse
250
Epilepsy, DM, COPD, CKD, PVD, CAD
43
2yrs
-
47
Same encounter
-
205
28 days
+
+
85
45
3yrs/5mo
++
+
1500
34
94
10yrs
+
+
1500
35
47
4yrs/9mo
+
+
2000
78
46
4yrs/3mo
+
289
Hypothyroid, HTN, AF, CAD, GERD, C diff, CKD.
500
44
3mo
+
Epilepsy, HTN, recurrent UTI
750
67
Unknown
-
Alzheimer’s dementia, MDD
HTN, hyperlipidemia, anemia, macular degeneration
500
18
56
Unknown
-
A A
Schizoaffective, SA
COPD
2000
74
69
1yr/9mo
+
78/ M
A A
Alzheimer’s dementia, schizophrenia
Epilepsy, HTN, pneumonia, UTI, neuropathy, anemia
1000
56
3yr/6mo
+
47/ M
A A
MDD, schizophrenia
Asthma, HTN
3000
43
67
6yr 1mo
+
+
47/F
C
Schizoaffective, ID
Epilepsy, GERD
2000
81
188
4yrs/3mo
-
+
45/F
C
BPD
HTN
1500
59
5yrs
-
1000
70
Same encounter
-
1500
57
6yrs/8mn
+
94
1yr/1mo
+
88/ M
C
Dementia
CHF, HTN, AF, CKD, asthma
1 7
53/ M
A A
Schizophrenia, MDD with SI
HTN, DM, SLE, TIA, lung CA
1 8
21/ M
A A
IED, ID
1000
80
1 51/ A BPD, SA 1500 79 Unknown 9 M A Same 2 67/F C Schizophrenia Asthma, hyperlipidemia 600 79 + 0 encounter 1 – Sex: M=male; F=female 2 – AA = African-American; C = Caucasian. 3 – In addition to discontinuing VPA, lactulose was the only other treatment used for VHE 4 – + = VPA continued;++ = VPA dosage increased - = VPA discontinued. 5 – + = lactulose used AF = atrial fibrillation; BPD = bipolar disorder; CA = cancer; CAD = coronary artery disease; C diff = clostridium difficile; CHF = congestive heart failure; CKD = chronic kidney disease; COPD = chronic obstructive pulmonary disease; DM = diabetes mellitus; GERD = gastroesophageal reflux disease; h/o = history of; HTN = hypertension; ID = intellectual disability; IED = intermittent explosive disorder; MDD = major depressive disorder; PVD = peripheral vascular disease; SA = substance abuse; SI = suicidal ideation; SLE = systemic lupus erythematosis; TIA = transient ischemic attack; UTI = urinary tract infection; VHE = valproate-associated hyperammonemic encephalopathy; VPA = valproic acid (or divalproex sodium)
12
PII: DOI: Reference:
S0033-3182(17)30041-5 http://dx.doi.org/10.1016/j.psym.2017.02.003 PSYM738
To appear in: Psychosomatics Cite this article as: Chandani Lewis, George E. Tesar and Roman Dale, Valproate-induced Hyperammonemic Encephalopathy (VHE) in General Hospital Patients with one or more Psychiatric DisordersValproate-induced hyperammonemic encephalopathy in a general hospital☆, Psychosomatics, http://dx.doi.org/10.1016/j.psym.2017.02.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ORIGINAL ARTICLE Valproate-induced hyperammonemic encephalopathy (VHE) in general hospital patients with one or more psychiatric disorders Chandani Lewis 1,2, George E. Tesar 1, Roman Dale 1 1
Dept. of Psychiatry and Psychology, Neurological Institute, Cleveland Clinic, Cleveland, OH 44195
USA 2
Current address: Medical Department, Zepf Center, 6605 W Central Ave, Toledo, OH 43615 & Dept.
of Psychiatry, University of Toledo College of Medicine and Life Sciences, Toledo, OH 43615, USA.
Address for Correspondence: Chandani Lewis, MD Medical Department Zepf Center, 6605 W Central Ave Toledo, OH 43615 419-383-5683 Phone: (419) 383-5683; Fax: (419) 383- 5630; Email: [email protected]
Short Title: Valproate-induced hyperammonemic encephalopathy in a general hospital
Conflicts of interest: Nothing to declare for all authors. Date:
1
Financial Disclosure No direct funding was received for this study. No specific salary was set aside or given for the writing of this paper. No funding bodies had any role in the study design, data collection, analysis, decision to publish or preparation of the manuscript.
Abstract Background: Divalproex sodium/valproic acid (VPA) is an antiepileptic drug approved for use in epilepsy and bipolar disorder. Valproate-induced hyperammonemia (VHA) occurs in up to 50 % of VPA-treated patients, some of which may become encephalopathic. Valproate-induced encephalopathy (VHE) is thought to be rare, and for a variety of reasons the diagnosis requires a high index of suspicion. The study’s goals are to determine how common VHE is, and the quality of treatment provided when diagnosed.
Methods: Retrospective, cross-sectional survey of general hospital patients. The hospital’s laboratory and pharmacy databases were combined to identify a cohort of all VPA-treated patients during a five-year period that developed hyperammonemia. Hospital records of the subset of patients with a psychiatric disorder were selected and reviewed for data collection. Results: Twenty of 793 VPA-treated patients (2.52%) had signs and symptoms consistent with VHE.
The majority were Caucasian males on multiple psychotropic agents. Valproate was
appropriately discontinued in eight (40%). Lactulose was the only ammonia-lowering drug used, and administered to six patients only one of who also had VPA discontinued.
2
Conclusion: Study results indicate that valproate-induced hyperammonemic encephalopathy (VHE) may be more common in psychiatric patients than previously assumed, but under-recognized and inadequately treated. The diagnosis of VHE requires a high-index of suspicion.
Outcome is
favorable once it is recognized and treated appropriately.
Introduction Valproate-induced hyperammonemic encephalopathy (VHE), recently reviewed by our group and another (1, 2), is recognized as a rare, but serious and unpredictable adverse outcome of treatment with valproic acid (VPA) including its congener, divalproex sodium. Most of the existing VHE literature is based on treatment of epilepsy (3, 4); there are fewer reports of its use for bipolar disorder or off-label indications such as anxiety, agitation, or delirium (2, 5). Clinical findings of VHE are nonspecific, and typically present acutely or subacutely as drowsiness, lethargy, agitation, confusion or coma. Other features include ataxia, vomiting, focal neurological signs, low grade fever, and seizures (1, 2, 5).
Recognition of VHE requires a high level of clinical suspicion, since clinical presentation is nonspecific and correlates poorly with dosage, blood levels or duration of VPA treatment (5). The transient, mild liver enzyme elevation often seen during VPA therapy is are not necessarily present, and, similar to hepatic encephalopathy, correlation of VHE severity and blood ammonia levels is variable (5). Asymptomatic ammonia elevations are seen in as many as 51% of VPA-treated psychiatric patients (6); in contrast, normal serum ammonia levels have been reported in unusual cases of VHE (5).
3
Our impression is that VHE may be more common than previously reported. To test this assumption we investigated a sample of general hospital psychiatric patients. In addition to frequency of occurrence of VHE, we were interested in its rate of recognition, and the quality of treatment provided when recognized. To our knowledge this is the first study to examine VHE frequency in a discrete patient population.
Methods This is a retrospective, case-based, cross-sectional survey of general hospital patients treated with divalproex sodium or valproic acid (VPA) during a five-year period, 2004 to 2009. The cohort of interest was patients with one or more psychiatric disorders who developed VHE. The list of study patients was generated by comparing the hospital’s pharmacy and laboratory databases to identify all VPA-treated patients who also had evidence of hyperammonemia during their index admission. In order to reduce the likelihood of false positives, hyperammonemia was defined as a blood ammonia level of 41 µmol/L or more (that is, 15% above the upper limit of normal, 35 µmol/L). Hyperammonemic patients with at least one primary psychiatric diagnosis were selected for detailed chart review.
Two investigators (CL and RD) reviewed clinical records for documented evidence of an acute change in mental status consistent with a diagnosis of encephalopathy/delirium. Key search terms included agitation, altered sensorium or mental status, cognitive deficits, confusion, lethargy, delirium, delusions, dementia, disorientation, falls, hallucinations, memory impairment, and speech abnormalities.
4
Patients with hepatic enzyme elevations (aspartate transaminase [AST] and alanine aminotransferase [ALT]) were excluded to eliminate hepatic failure as a cause of hyperammonemia.
This study was approved by the Cleveland Clinic Institutional Review Board.
Results
During the five-year study period the hospital’s pharmacy database listed 793 VPA-treated inpatients; the laboratory database included 294 patients with one or more instances of hyperammonemia (blood ammonia level > 41 µmol/L). Pooling these data identified 30 VPA-treated patients with at least one report of hyperammonemia. Ten of these patients were excluded from further study for a variety of reasons, including elevated AST and ALT blood levels (n=7), seizure disorder alone with no associated psychiatric diagnosis (n=2), and insufficient evidence of encephalopathy (n=1). Detailed record review was performed in the remaining 20.
Demographic, diagnostic, treatment, and laboratory data are listed in the Table. The average age of the patients was 59.55 years+19.35 (range, 21-89). Fifteen (75%) were male, 14 were Caucasian (70%), and six were African American (30%). Twelve (60%) were admitted to the general medicine service and eight were admitted to either an adult (n=4) or geriatric inpatient psychiatry unit (n=4).
The 20 patients collectively had 31 psychiatric diagnoses including schizophrenia or schizoaffective disorder (n=9), dementia (n=6), major depressive disorder (n=5), bipolar disorder (n=4), current or a history of substance abuse disorder (n=4), intellectual disability (n=3), and intermittent explosive
5
disorder (1). Seven had a single psychiatric diagnosis, including two who also had a diagnosis of intellectual disability.
The most common medical co-morbidities were hypertension (n=11), diabetes mellitus (n=5), epilepsy (n=5), cardiovascular or peripheral vascular diseases (4), and chronic kidney disease (3).
The average number of psychoactive drugs per patient was 2.8+1.39 (range 1-6). In addition to VPA, patients received other mood stabilizers: oxcarbazepine (n=3), lithium (n=3), lamotrigine (n=2), and topiramate (n=1); antipsychotics: quetiapine (n=7), risperidone (n=3), aripiprazole (n=2), ziprasidone (n=2), olanzapine, haloperidol and clozapine, (n=1 each); antidepressants: serotonin selective reuptake inhibitors (n=8) and mirtazapine (n=2); or anxiolytics: benzodiazepines (n=6), and gabapentin (n=3).
The mean daily VPA dose was 1280 mg+696.5 (range, 250-3000 mg per day). Blood levels of VPA were available in 10 patients with a mean level of 81.7µg/l+ 6.79 (range, 18 –289µg/l). All VPA blood levels were within therapeutic range except in the patient who overdosed (289µg/l).
Ammonia levels were drawn in the morning in all but three patients with blood levels ranging from 43205 (median, 63); only two patients – one of who overdosed on VPA – had serum ammonia levels > 100 µmol).
Hyperammonemia was first detected after a variable period following initiation of VPA treatment: within the first month (n=4); within one year (n=1); between one and five years (n=9); between six to 10 years (n=3); and for an unknown duration (n=3).
6
In addition to typical signs and symptoms of encephalopathy, two patients were diagnosed with presumptive treatment-resistant seizures.
Despite evidence of hyperammonemia and encephalopathy 12 patients (60%) continued to receive VPA, including one whose dosage was increased. Lactulose was administered to six patients (30%) only one of who also stopped VPA. No patient received alternative agents indicated for treatment of hyperammonemia (e.g., levo-carnitine, charcoal, neomycin, or rifamixin). .
Discussion
Our study suggests that the occurrence of VHE in psychiatric patients is by no means rare, and yet often overlooked. In the population studied, 20 of 793 VPA-treated patients (2.52%) developed retrospectively diagnosed VHE. This may in fact underestimate the true frequency of VHE given that hyperammonemia rather than a diagnosis of delirium or encephalopathy was the main selection criterion for study participation. This methodological limitation over-looks VPA-treated patients who may have developed VHE but had either a normal ammonia level, an ammonia level between 36 to 40 µmol, or did not have blood ammonia checked at all. Gerstner and colleague’s meta-analysis of 51 cases of VHE reported during a 10 year period in patients with epilepsy (7) may provide a point of comparison. Assuming a rate of VPA-treatment similar to our group the average number of cases per year identified by Gerstner et al, which is, five, is nearly the same as the four per year identified in our study .
7
Although the study’s methodology prevents firm conclusions regarding correlations between VHE and a variety of variables, including age, gender, ammonia levels and duration or intensity of VPA treatment, our results are largely consistent with previous reports.
In this exclusively adult population, VHE occurred across the life-span (ages 21- 89) with seven patients (35%) aged 65 or older. Since other age-related variables increase the likelihood of preexisting cognitive impairment, VPA-treated patients older than 65 should have blood ammonia levels checked both before and after starting VPA treatment. Accurate interpretation of ammonia blood levels in the elderly, however, should be informed by a recent study reporting a two-fold increase in serum ammonia levels in older psychiatric patients (8).
The duration of VPA therapy varied from a few days to 10 years. Among the 10 patients with documented VPA blood levels, all were within normal range except for one from the patient who overdosed on VPA. The apparent lack of correlation between VPA blood levels and hyperammonemia or the likelihood of developing VHE supports previous reports (3-5, 9)
Only two of our patients (10%) had ammonia levels exceeding 100 µmols/L, which parallels previous studies showing poor correlation between ammonia level and VHE presence or severity. Both Dealberto et al (5) and Verrotti et al (4) reported a positive correlation between duration of VPA therapy and serum ammonia levels, but, as noted, ammonia level is a poor predictor of clinical presentation. .
8
Patients in our study were on an average of three psychoactive drugs with a maximum number of six. Polypharmacy (> 3 agents) has been identified as a risk factor for developing hyperammonemia (2, 5) and may increase the risk of VHE by interfering with normal urea cycle activity, reducing carnitine levels, or displacing highly protein-bound VPA (1). Simultaneous use of VPA and other drugs reported to cause VHE (e.g., topiramate, lamotrigine) or pharmacokinetic interaction with VHA (e.g., risperidone) should be avoided or monitored carefully (10-13).
Detection and management of VHE in our population was suboptimal similar to the experiences reported by others (2, 5). Diagnosis of VHE warrants either discontinuation of VPA or dosage reduction depending on symptom severity (6). Treatment with VPA was discontinued in only eight of our patients (40%) and the dosage was increased in one patient underscoring the importance of maintaining a high index of suspicion for VHE.
Presenting as either hypoactive or hyperactive
delirium, VHE can be confused with depression or manic agitation potentially contributing to the misinterpretation of inadequate VPA-treatment (6, 14). While severe VHE can present with epileptic seizures (4) it is more likely to present with non-epileptic events either misdiagnosed as epileptic or as an exacerbation of a primary psychiatric disorder. Any of these circumstances can lead to inappropriate escalation rather than de-escalation of VPA-treatment. For example, two of our patients (case 4 & 5) with “uncontrolled seizures” continued to receive VPA when it should have been discontinued.
Persistence of VHE despite reduction or discontinuation of VPA is an indication for additional ammonia–depleting agents such as lactulose, charcoal, neomycin, rifamixin, or levo-carnitine (6, 15). Five patients (30%) received any kind of treatment, which in all instances was lactulose. None of our patients required hemodialysis (indicated for ammonia blood levels >400 µmol/L) (16). There were no
9
reported deaths and all of our patients were discharged from the hospital similar to findings in other reports (5, 17).
Interpretation of our findings must be considered with the following study limitations in mind. The cross-sectional, small sample size, retrospective design, as previously noted, prevented firm conclusions about predictors of VHE. Additionally, the lack of a gold-standard diagnostic instrument or a clinical finding pathognomonic of VHE compromises diagnostic validity thereby increasing the likelihood of either false-positive or false-negative diagnoses. As discussed previously, however, the likelihood of false-negatives (under-diagnosis) is probably greater than false-positives (over-diagnosis) supporting the contention that true VHE frequency is likely greater than the 2.52% identified in our population. Availability of EEG recordings may have helped improve diagnostic validity, but it was unlikely to facilitate differentiation of VPA from other causes of encephalopathy.
Systematic
assessment of ammonia levels in all VPA-treated patients including those who did not develop VHE prevented firm conclusions about the relationship between VHE and blood ammonia levels.
Conclusion: During a five-year study interval valproate-induced hyperammonemic encephalopathy (VHE) was identified in 2.52% of valproate (VPA)-treated general hospital psychiatric patients. Study methodology very likely contributed to under-estimation of the true frequency of VHE and only a minority of patients with VHE received adequate treatment. Our findings indicate a need for increased familiarity with the diagnosis and with the appropriate treatment of VHE.
10
References: 1. Lewis C, Deshpande A, Tesar GE, Dale R. Valproate-induced hyperammonemic encephalopathy: a brief review. Curr Med Res Opin. 2012;28(6):1039-42. 2. Chopra A, Kolla BP, Mansukhani MP, Netzel P, Frye MA. Valproate-induced hyperammonemic encephalopathy: an update on risk factors, clinical correlates and management. Gen Hosp Psychiatry. 2012;34(3):290-8. 3. Segura-Bruna N, Rodriguez-Campello A, Puente V, Roquer J. Valproate-induced hyperammonemic encephalopathy. Acta Neurol Scand. 2006;114(1):1-7. 4. Verrotti A, Trotta D, Morgese G, Chiarelli F. Valproate-induced hyperammonemic encephalopathy. Metab Brain Dis. 2002;17(4):367-73. 5. Dealberto MJ. Valproate-induced hyperammonaemic encephalopathy: review of 14 cases in the psychiatric setting. Int Clin Psychopharmacol. 2007;22(6):330-7. 6. Raja M, Azzoni A. Valproate-induced hyperammonaemia. J Clin Psychopharmac. 2002;22(6):631-3. 7. Gerstner T, Buesing D, Longin E, Bendl C, Wenzel D, Scheid B, et al. Valproic acid induced encephalopathy--19 new cases in Germany from 1994 to 2003--a side effect associated to VPA-therapy not only in young children. Seizure. 2006;15(6):443-8. 8. Adler LW, Regenold WT. Valproate-Related Hyperammonemia in Older Adult Psychiatric Inpatients. Prim Care Companion CNS Disord. 2015;17(2). 9. Mehndiratta MM, Mehndiratta P, Phul P, Garg S. Valproate induced non hepatic hyperammonaemic encephalopathy (VNHE)--a study from tertiary care referral university hospital, north India. JPMA J Pak Med Assoc. 2008;58(11):627-31. 10. Latour P, Biraben A, Polard E, Bentue-Ferrer D, Beauplet A, Tribut O, et al. Drug induced encephalopathy in six epileptic patients: topiramate? valproate? or both? Hum Psychopharmacol. 2004;19(3):193-203. 11. Kimmel RJ, Irwin SA, Meyer JM. Valproic acid-associated hyperammonemic encephalopathy: a case report from the psychiatric setting. Int Clin Psychopharmacol. 2005;20(1):57-8. 12. Reif A, Leonhard C, Mossner R, Lesch KP, Fallgatter AJ. Encephalopathy and myoclonus triggered by valproic acid. Prog Neuropsychopharmacol Biol Psychiatry. 2004;28(6):1061-3. 13. Carlson T, Reynolds CA, Caplan R. Case report: valproic Acid and risperidone treatment leading to development of hyperammonemia and mania. J Am Acad Child Adolesc Psychiatry. 2007;46(3):356-61. 14. Carr RB, Shrewsbury K. Hyperammonemia due to valproic acid in the psychiatric setting. Am J Psychiatry. 2007;164(7):1020-7. 15. Lheureux PE, Hantson P. Carnitine in the treatment of valproic acid-induced toxicity. Clin. Toxicol. 2009;47(2):101-11. 16. Eyer F, Felgenhauer N, Gempel K, Steimer W, Gerbitz KD, Zilker T. Acute valproate poisoning: pharmacokinetics, alteration in fatty acid metabolism, and changes during therapy. J Clin Psychopharmacol. 2005;25(4):376-80. 17. Raby WN. Carnitine for valproic acid-induced hyperammonemia. Am J Psychiatry. 1997;154(8):1168-9.
11
TABLE. Demographic, diagnostic, laboratory measures and treatment in 20 patients with VHE C as e
Age/ Sex1
Ra ce2
Psychiatric diagnosis
Medical Co-morbidities
VPA dose (max mg/d)
VPA blood level (µg/l)
Ammonia serum level (µmol)
VPA duration (yr/mo)
VHE Treatmen t3 Lac VP tuA4 lose 5
1
89/F
C
MDD, Alzheimer’s Dementia
2
44/ M
C
BPD
DM, HTN
500
C
Paranoid schizophrenia
DM, HTN
1000
C
BPD
DM, HTN, Epilepsy
2000
3 4
60/ M 52/ M
5
34/ M
C
MDD, SA, personality disorder
6
64/ M
C
Schizophrenia, ID
7
44/ M
C
Schizoaffective
8
89/ M
C
9
67/ M
C
1 0
87/ M
C
1 1
64/F
1 2 1 3 1 4 1 5 1 6
Alzheimer’s dementia, psychosis Dementia, h/o alcohol abuse
250
Epilepsy, DM, COPD, CKD, PVD, CAD
43
2yrs
-
47
Same encounter
-
205
28 days
+
+
85
45
3yrs/5mo
++
+
1500
34
94
10yrs
+
+
1500
35
47
4yrs/9mo
+
+
2000
78
46
4yrs/3mo
+
289
Hypothyroid, HTN, AF, CAD, GERD, C diff, CKD.
500
44
3mo
+
Epilepsy, HTN, recurrent UTI
750
67
Unknown
-
Alzheimer’s dementia, MDD
HTN, hyperlipidemia, anemia, macular degeneration
500
18
56
Unknown
-
A A
Schizoaffective, SA
COPD
2000
74
69
1yr/9mo
+
78/ M
A A
Alzheimer’s dementia, schizophrenia
Epilepsy, HTN, pneumonia, UTI, neuropathy, anemia
1000
56
3yr/6mo
+
47/ M
A A
MDD, schizophrenia
Asthma, HTN
3000
43
67
6yr 1mo
+
+
47/F
C
Schizoaffective, ID
Epilepsy, GERD
2000
81
188
4yrs/3mo
-
+
45/F
C
BPD
HTN
1500
59
5yrs
-
1000
70
Same encounter
-
1500
57
6yrs/8mn
+
94
1yr/1mo
+
88/ M
C
Dementia
CHF, HTN, AF, CKD, asthma
1 7
53/ M
A A
Schizophrenia, MDD with SI
HTN, DM, SLE, TIA, lung CA
1 8
21/ M
A A
IED, ID
1000
80
1 51/ A BPD, SA 1500 79 Unknown 9 M A Same 2 67/F C Schizophrenia Asthma, hyperlipidemia 600 79 + 0 encounter 1 – Sex: M=male; F=female 2 – AA = African-American; C = Caucasian. 3 – In addition to discontinuing VPA, lactulose was the only other treatment used for VHE 4 – + = VPA continued;++ = VPA dosage increased - = VPA discontinued. 5 – + = lactulose used AF = atrial fibrillation; BPD = bipolar disorder; CA = cancer; CAD = coronary artery disease; C diff = clostridium difficile; CHF = congestive heart failure; CKD = chronic kidney disease; COPD = chronic obstructive pulmonary disease; DM = diabetes mellitus; GERD = gastroesophageal reflux disease; h/o = history of; HTN = hypertension; ID = intellectual disability; IED = intermittent explosive disorder; MDD = major depressive disorder; PVD = peripheral vascular disease; SA = substance abuse; SI = suicidal ideation; SLE = systemic lupus erythematosis; TIA = transient ischemic attack; UTI = urinary tract infection; VHE = valproate-associated hyperammonemic encephalopathy; VPA = valproic acid (or divalproex sodium)
12