To the Editor:

Increased synthesis of fetal hemoglobin may ameliorate the clinical severity of sickle cell disease1; both chemotherapeutic and nonchemotherapeutic agents have been shown to augment fetal hemoglobin synthesis in selected populations of affected patients.2,3 Some studies have suggested that the fatty acid analogue valproic acid (n-dipropylacetic acid) may increase the synthesis of fetal hemoglobin. A study of 36 patients with epilepsy treated with valproic acid (10 to 46 mg/kg body weight per day) reported that the percentage of red blood cells containing fetal hemoglobin (F cells) was significantly higher in these individuals than in 293 patients not receiving this therapy; parallel changes in fetal hemoglobin were not reported.4 A study of four adults with sickle cell disease treated with valproic acid (15 to 40 mg/kg/d) reported a threefold increase in fetal hemoglobin in three patients, with no corresponding decrease in the frequency of vaso-occlusive crises, over 2 to 13 weeks.5 Finally, a study in which 10 patients with sickle cell disease received treatment with hydroxyurea followed by valproic acid (20 mg/kg/d) reported no difference in the concentrations of fetal hemoglobin, frequency of vaso-occlusive crises, or adverse drug effects in patients treated with either agent.6 

We examined the changes in fetal hemoglobin synthesis in individuals without sickle cell disease and receiving valproic acid for epilepsy to determine whether previously reported valproic acid-induced changes in F-cells are associated with measurable increases in fetal hemoglobin. In parallel, we evaluated laboratory and clinical responses associated with valproic acid therapy in seven patients with sickle cell disease.

One hundred and six consecutive patients attending the Neurology Clinic at The Hospital for Sick Children, Toronto, with a diagnosis of seizure disorder, none of whom had sickle cell disease or trait, were studied. Fifty-seven patients, aged (mean ± SD) 9.3 ± 4.4 years, were receiving therapy with carbamazepine (Tegretol), whereas 49 patients, aged 9.9 ± 5.2 years, had been treated with valproic acid (15 to 50 mg/kg/d) for 4.0 ± 0.5 years. Although a significant mean increase in mean red blood cell volume (from 84 ± 5 fL before treatment with valproic acid to 90 ± 5 fL; P < .01) had been observed in the latter group, the percentage of fetal hemoglobin (mean ± SEM, 0.9% ± 0.1%) in patients with therapeutic serum drug concentrations of valproic acid did not differ significantly from that of the carbamazepine-treated patients (0.8% ± 0.1%; P = NS).

In parallel, seven patients, aged 16.5 ± 10 years, with severe sickle cell disease, defined as three or more hospitalizations in the year before treatment, were offered valproic acid at 15 ± 3 (range, 9 to 20) mg/kg/d. Complete blood counts were obtained twice monthly. Patients were reviewed by a physician twice a month and questioned regarding adverse effects of treatment, sickle cell disease-related pain, admissions to hospital, and compliance with therapy. Annual rates of vaso-occlusive crisis were calculated as previously described by dividing the number of crises by the number of years of therapy (for example, 2 crises in 6 months = 4 crises per year).3 The number of days spent in hospital and the number of red blood cell transfusions administered were calculated in a similar manner. Compliance with valproic acid was monitored by the Medication Event Monitoring System, a bottle with a computer chip in the lid that determines the timing and frequency of bottle openings,7 and by determinations of concentrations of serum valproic acid obtained at clinic visits.

The changes observed during therapy with valproic acid over 5.8 ± 0.9 (3.2 to 9.0) months are shown in Table 1. Data are presented as the mean ± standard deviation.

Compliance with valproic acid was 93% ± 7% drug taken of that prescribed. Mean trough serum levels (561 ± 115 μmol/L) were within the therapeutic range (350 to 750 μmol/L; 50 to 100 μg/mL). No changes in liver function tests were observed; in one patient, the platelet count decreased to 98 × 109/L, prompting discontinuation of valproic acid. Review of this patient's medical records showed that he had chronic low-grade thrombocytopenia before the initiation of valproic acid; this finding has persisted after withdrawal of valproic acid. Bone marrow aspiration showed abundant megakarocytes. The relationship between thrombocytopenia in this patient and valproate therapy, acknowledged to induce thrombocytopenia in 21% to 60% of treated patients who have serum concentrations exceeding 100 to 140 μg/mL,8 9 is unclear.

Because of the lack of clinical improvement and frequent hospital admissions during valproic acid, one patient requested discontinuation of this therapy and initiation of hydroxyurea. In this patient, the percentage of fetal hemoglobin increased from a value of 9% immediately after 5 months of therapy with valproic acid to 24% after 11 months of treatment with hydroxyurea (Fig 1); after the initiation of hydroxyurea, this patient required no further admissions to hospital.

Fig. 1.

Shown are concentrations of fetal hemoglobin in a patient with sickle cell disease treated initially with valproic acid over 5 months and subsequuently with oral hydroxyurea at an initial dose of 11 mg/kg/d (arrows). Fetal hemoglobin at the start of valproic acid therapy was 8%, increasing to 9% during treatment. The patient discontinued valproic acid and began hydroxyurea. During the subsequent 11 months of hydroxyurea therapy, a steady increase in fetal hemoglobin was noted; the present value of fetal hemoglobin, during therapy with 16.2 mg of hydroxyurea per kilogram body weight per day, is 23.7%.

Fig. 1.

Shown are concentrations of fetal hemoglobin in a patient with sickle cell disease treated initially with valproic acid over 5 months and subsequuently with oral hydroxyurea at an initial dose of 11 mg/kg/d (arrows). Fetal hemoglobin at the start of valproic acid therapy was 8%, increasing to 9% during treatment. The patient discontinued valproic acid and began hydroxyurea. During the subsequent 11 months of hydroxyurea therapy, a steady increase in fetal hemoglobin was noted; the present value of fetal hemoglobin, during therapy with 16.2 mg of hydroxyurea per kilogram body weight per day, is 23.7%.

Close modal

These data show that, during valproic acid therapy in patients without sickle cell disease, previously reported increases in F-cells4 are not reflected by measurable changes in the synthesis of fetal hemoglobin. During valproic acid therapy in patients with sickle cell disease, a modest mean increase in fetal hemoglobin concentration over 6 months was not paralleled by changes in crisis rate, number of transfusions required, or total days of hospitalization. The lack of clinical effect in these patients stands in contrast to that observed in a group of patients of comparable age treated with hydroxyurea.10-12 Also of striking contrast to the modest hematologic responses of these patients during treatment with valproic acid are the changes in clinical status and fetal hemoglobin concentration observed in one patient during subsequent hydroxyurea therapy, which induced a near tripling of fetal hemoglobin from the values attained during valproate therapy.

Although all of these patients had serum concentrations of valproic acid within the therapeutic range for antiseizure effect, they received slightly lower daily doses of valproic acid than that prescribed in previous studies.4-6 Although higher daily doses of valproic acid may more substantially increase fetal hemoglobin synthesis in patients with sickle cell disease, heightened risks of hematologic or hepatic toxicity may accompany more intensive dosing regimens of this agent.8,9 13-15 

We thank Susan Scorizzi for technical assistance and Dr Graham Sher for supervising the management of some of the patients. Dr Olivieri is a Scientist of the Medical Research Council of Canada.

1
Embury
SH
The clinical pathophysiology of sickle cell disease.
Annu Rev Med
37
1986
361
2
Dover GJ, Charache S: Increasing fetal hemoglobin production in sickle cell disease: Results of clinical trials, in Stamatoyannopoulos G, Nienhuis AW (eds): Developmental Control of Globin Expression. New York, NY, Liss, 1987, p 456
3
Charache
S
Terrin
ML
Moore
RD
Dover
GJ
Barton
FB
Eckert
SV
McMahon
RP
Bonds
DR
and the Investigators of the Multicenter Study of Hydroxyurea in Sickle Cell Anemia
Effect of hydroxyurea on the frequency of painful crises in sickle cell anemia.
N Engl J Med
332
1995
1317
4
Collins
AF
Luban
NLC
Dover
GJ
Increased fetal hemoglobin production in patients receiving valproic acid for epilepsy (letter).
Blood
84
1994
1690
5
Wood A, Davies SC: Induction of haemoglobin F in sickle cell anaemia by sodium valproate. Blood 86:141a, 1995 (abstr, suppl 1)
6
Salim KA, Salim A: Therapy with sodium valproate is as effective as that with hydroxyurea in preventing painful crises of sickle cell disease. Blood 86:141a, 1995 (abstr, suppl 1)
7
Cramer
JA
Mattson
RH
Prevey
ML
Scheyer
RD
Ouellette
VL
How often is medication taken as prescribed? A novel assessment technique.
JAMA
261
1989
3273
8
Delgado
MR
Riela
AR
Mills
J
Browne
R
Roach
ES
Thrombocytopenia secondary to high valproate levels in children with epilepsy.
J Child Neurol
9
1994
311
9
May
RB
Sunder
TR
Hematologic manifestations of long-term valproate therapy.
Epilepsia
34
1993
1098
10
Scott
JP
Hillery
CA
Brown
ER
Misiewicz
VM
Labotka
RJ
Hydroxyurea therapy in children severely affected with sickle cell disease.
J Pediatr
128
1996
820
11
Ferster
A
Vermylen
C
Cornu
G
Buyse
M
Corazza
F
Devalck
C
Fondu
P
Toppet
M
Sariban
E
Hydroxyurea for treatment of severe sickle cell anemia: A pediatric clinical trial.
Blood
88
1996
1960
12
Jayabose
S
Tugal
O
Sandoval
C
Patel
P
Puder
D
Lin
T
Visintainer
P
Clinical and hematologic effects of hydroxyurea in children with sickle cell anemia.
J Pediatr
129
1996
559
13
Willmore
LJ
Wilder
BJ
Bruni
J
Villarreal
HJ
Effect of valproic acid on hepatic function.
Neurology
28
1978
961
14
Sussman
NM
McLain
LW
A direct hepatotoxic effect of valproic acid.
JAMA
242
1979
1173
15
Gidal
B
Spencer
N
Maly
M
Pitterle
M
Williams
E
Collins
M
Jones
J
Valproate-mediated disturbances of hemostasis: Relationship to dose and plasma concentration.
Neurology
44
1994
1418
Sign in via your Institution