вход по аккаунту


Carbamazepine and hematological monitoring.

код для вставкиСкачать
Carbamazepine and Hematological Monitoring
Robert G. Hart, MD, and J. Donald Easton, MD
Clinically important hematological toxicity is uncommon with carbamazepine (CBZ). Only 20 patients with CBZassociated aplastic anemia have been reported since 1%.
Leukopenia and thrombocytopenia occur in about 2% of
patients. Based on the available data on hematological complications of CBZ, we question the justification for
expensive, excessive laboratory monitoring.
Hart RG, Easton JD: Carbamazepine and hematological monitoring. Ann Neurol 11:309-3 12, 1982
In a recent report the yearly cost of carbamazepine
(CBZ) administration (1,000 mg per day), exclusive
of physician’s fees, was estimated to be $978, of
which $248 was for medication and $730 was for laboratory tests [71. This estimate for laboratory cost is
conservative. Frequent hematological monitoring is
recommended during treatment, and includes complete blood and platelet counts before treatment, at
weekly intervals during the first month of treatment,
every two weeks for the next two to three months,
and monthly thereafter [5, 11,23,30]. At our institution these tests would cost $745 in the first year and
$497 yearly thereafter. Abiding by the manufacturer’s 1980 recommendations,* the first-year cost
would be $2,161. Estimating that 200,000 patients in
the United States are taking CBZ, annual laboratory
costs would exceed $300 million! (Gibney P, CibaGeigy Corporation, Pharmaceutical Division, Summit, NJ: personal communication, 1981).
The rationale for this recommended expenditure
needs reexamination in light of the reported experience with CBZ. Additionally, a poll of neurologists
in our area showed that none abided by even the
minimal recommendations just described. Therefore,
we present a review of the data on hematological
monitoring and CBZ.
The justification for laboratory monitoring of these
complications rests on three issues:
1. Does the prevalence of toxic reactions justify laboratory monitoring? Many medications other than
CBZ are occasionally associated with bone marrow suppression, but rarity precludes routine
monitoring (e.g., chloramphenicol, phenothiazines, phenytoin) [231.
2. Does detection of hematological toxicity by
routine monitoring alter the outcome, either by
early withdrawal of CBZ or by specific treatment?
3. If the first two questions are answered in the
affirmative, is the currently recommended monitoring regimen optimal for detecting serious
reactions? Since aplastic anemia can occur suddenly after months of therapy, why is a monthly
rather than weekly or bimonthly regimen chosen
[ 18]? The current recommendations for laboratory monitoring are based on De Vries’ 1965 recommendations for monitoring of all anticonvulsants [31.
+Completeblood, platelet, and reticulocyte count and serum iron
determination before treatment, weekly for three months, then
monthly, along with liver enzymes, blood urea nitrogen urinalysis,
and ophthalmological evaluation at regular intervals [ 5 ] (assumed
to be bimonthly for the cost calculation given here).
Aplastic Anemia
Fatal aplastic anemia is the most feared complication
of CBZ therapy. Pisciotta [231 reviewed 13 cases
reported prior to 1975. Because of multiple-drug
therapy and additional factors, CBZ was the probable
cause in only 3 of the patients. Others have doubted
the existence of a causal relationship between CBZ
and aplastic anemia [ 161. The Geigy Corporation has
since received 6 further reports of aplastic anemia,
and all 6 patients were on multiple-drug therapy
(Gibney P: personal communication, 1981). Despite
the widespread use of this agent for trigeminal
neuralgia and epilepsy, the number of known cases of
CBZ-associated aplastic anemia is only 20 (13 fatal)
([17, 231 and Gibney P: personal communication,
From the Department of Neurology, University of MissouriColumbia School of Medicine, Columbia, MO 65212.
Received Apr 29, 1981, and in revised form July 13. Accepted for
publication July 19, 1981.
Hematological Toxicity
The serious hematological toxicity of CBZ consists
of rare aplastic anemia (pancytopenia), persistent
leukopenia, and isolated thrombocytopenia (Table).
Address reprint requests to Dr Hart.
0364-5 134/82/030309-04$01.25 @ 1981 by the American Neurological Association 303
Hematological Toxicity of Cadamazepine
Aplastic anemia
Case reports: 20 ([17, 231 and Gibney P: personal
communication, 198 1)
Prevalence: < 1/50,000
Incidence: 0.5/100,000/yr
Leukopenia (neutropenia)
Transient: 10% (range, 2-60%) [2, 8, 10, 12-15,
17, 21, 24, 26, 321
Persistent: 2c; (range, 0 - 8 c i ’ ) 16, 8, 11, 12, 15, 19,
26, 321
Average change“: 0 [ I , 11, 12, 19, 27,281 to 1,000
per mm3 126, 321
Prevalence: 256 [2,11, 14, 301
Average changea: 0 [14, 321 to 20,000 per mrn3 [ l ,
24, 281
Prevalence: <5‘; .(range, 0-10“1) 113, 19, 291
Average change”: 0 11, 14, 19, 27, 281 to 0.5 gm Hb/dl
”Average change is the average decrease for all patients in the report.
1981). It is difficult to estimate how many cases of
CBZ-associated aplastic anemia go unrecognized or
unreported, but they must be rare. Among 225
epileptic patients reported by Livingston et a1 [15]
who took CBZ for an average of 40 months, no serious hematological toxicities occurred (see next section). Many smaller series of epileptic patients include
no instances of aplastic anemia [26-29, 321. The duration of CBZ therapy prior to the onset of aplastic
anemia varied fairly evenly between 4 and 330 days,
and, contrary to early reports, patient age and total
dose do not appear to be major factors in the development of aplastic anemia ([23] and Gibney P:
personal communication, 1981).
Pisciotta [23] stated “with respect to chloramphenicol, where the incidence of marrow suppression
was calculated . . . to be one i n 20,000 to one in
40,000, the extreme rarity of [aplastic anemia] precludes its identification by routine blood counts.”
Experience suggests that this complication may be
even rarer with CBZ than with chloramphenicol. Pisciotta also reported 5 cases of agranulocytosis among
6,500 patients receiving phenothiazines who were
monitored by weekly blood counts from weeks 2
through 10. He concluded: “We are not sure that all
this effort was worthwhile and in fact no longer d o
screening tests.” This is despite a lower mortality
when agranulocytosis was identified by laboratory
monitoring in patients receiving phenothiazines
compared to phenothiazine-treated patients who developed infection before agranulocytosis was diag-
310 Annals of Neurology
Vol 11 No 3 March 1782
nosed [231. O n the basis of the known cases of aplastic anemia associated with CBZ, routine laboratory
monitoring is not justified when compared to the
situation with other potentially myelotoxic drugs such
as phenothiazines and chloramphenicol.
Leukopenia (including neutropenia) occurs more
frequently than aplastic anemia with CBZ therapy
(see the Table). Transient leukopenia, which resolves
despite continuation of CBZ, occurs in approximately 10% of patients, usually in the first month of
treatment [2, 11, 12, 15,241. Patients with a younger
average age are reported to have a lower prevalence
of transient leukopenia ( 4 % ) [8, 10, 13, 261 than
older patients (12%) [2, 12, 291, although dose,
monitoring regimen, and criteria of leukopenia
among the various reports are not constant. No discernible relationship exists between average dose of
CBZ and leukopenia. Mild, transient leukopenia is
more common in patients with low or low-normal
white blood cell (WBC) counts before treatment [ 1,
11, 121.
Livingston et a1 [ 151 reported moderate leukopenia
(3,000 to 3,500 WBC/mm3) in 9 of 225 patients and
marked leukopenia (2,000 to 3,000 WBC/mm3) in 4
of 225 patients, usually within the first three months
of treatment. All cell counts returned to normal despite continuation of CBZ, and in no case did isolated
leukopenia precede more widespread marrow suppression [ 151.
Persistent leukopenia has been reported in 0 to
8% of patients treated with CBZ (see the Table). In 2
recent case reports, profound leukopenia with attendant infection developed after one month of CBZ
treatment [6, 91. Extenuating factors (renal failure,
radiotherapy, and immunosuppression) cloud the
issue in 1 case [6]. Discontinuation of CBZ resulted
in return of the cell count to normal and clinical improvement. Substantial persistent leukopenia is usually evident within the first few weeks of therapy and
responds to discontinuation of CBZ [6, 9, 11, 15,
191. Less frequently, it may appear after several
months of treatment [2, 11, 12 291. It is uncertain
whether the leukopenia is dose related, although it
was proved to be so in 1 case [l 11. Leukocytosis associated with CBZ has been reported [20].
Thrombocylopen ia
Isolated thrombocytopenia is uncommon, although a
16-year-old girl was recently reported to develop
petechiae, with a platelet count of 1l,000/mm3, after
two weeks of CBZ therapy [25]. The platelet count
returned to normal after discontinuation of CBZ. In
another case, asymptomatic thrombocytopenia of
50,000/mm3 was detected on routine laboratory
monitoring after ten months of CBZ therapy, again
with recovery on cessation of the drug [22]. The authors speculated about whether the thrombocytopenia would have progressed to aplastic anemia
had it not been discovered on monthly blood cell
counts [22]. Mild, transient falls in platelet count
have been reported in the initial month of CBZ
therapy [ l , 24, 301.
Mild anemia has been reported in patients on chronic
CBZ therapy, with return to normal on cessation of
the drug [ 19, 321. Mild eosinophilia is common in the
initial weeks of treatment 121.
Hematological Monitoring
Serious hematological toxicity with CBZ is rare,
perhaps more so than with other drugs for which
routine laboratory monitoring is not carried out. We,
and others [13, 261, agree with Cereghino et a1 [l]
that “initial fears about toxicity, particularly bone
marrow depression, seem exaggerated in view of subsequent experience.”
Despite more widespread use of CBZ in the past
decade, there have been fewer reports of bone marrow toxicity ([13] and Gibney P: personal communication, 1981).The reason is unclear. Perhaps it is due
to less willingness to report this now recognized
complication. Alternatively, laboratory monitoring
may be helping by identifying patients with pretreatment hematological abnormalities so that CBZ
therapy is not begun (the majority of patients with
aplastic anemia had no pretreatment studies) or by
permitting early detection of bone marrow suppression so that therapy can be stopped before the patient
progresses to aplastic anemia. The manufacturer [ 5 ]
states that “early detection of hematologic change is
important since, in some patients, aplastic anemia is
reversible.” However, no evidence exists that early,
presymptomatic detection of marrow suppression by
means of laboratory monitoring alters the outcome.
Among the last 6 patients with aplastic anemia,
there are no data to indicate whether the complication was discovered by laboratory monitoring or by
the development of symptoms or signs (Gibney P:
personal communication, 1981). Notably, the mortality in this group was 33% ( 2 of 6) at a time when
monitoring was widespread, compared to 78% (11
of 14) in the earlier reports that antedated routine
monitoring ([23] and Gibney P: ,personal communication, 1981). Awareness of this complication, as
well as supportive treatment of aplastic anemia, has
changed over the interval.
Little information is available concerning persis-
tent leukopenia or isolated thrombocytopenia as a
prelude to more widespread, irreversible marrow
suppression, even though Case 3 of Pisciotta [23]
hints at leukopenia and thrombocytopenia preceding
pancytopenia. The data of Livingston et a1 suggest
that leukopenia, even when marked, is not a harbinger of aplastic anemia [ 151.
Recommendations for Laboratory Monitoring
Even assuming that the prevalence of serious bone
marrow toxicity justifies the effort of its detection
and that early detection improves the outcome, the
question of the optimal laboratory monitoring regimen remains. Almost certainly, inclusion of serum
iron assay and reticulocyte counts, as recommended
by the manufacturer, adds excessive (double) cost
and little benefit. Any regimen must take into account the clustering of leukopenia early in the course
of therapy, while the onset of aplastic anemia is
spread fairly evenly over the first year.
With annual laboratory costs for recommended
monitoring costing some $300 million, the current
rationale begs urgent evaluation and justification. A
major discrepancy exists between the manufacturer’s
recommendations and the routine practice of most
neurologists. Until more information is available
about the hematological toxicity of CBZ, we recommend the following:
1. Reports of serious hematological complications of
CBZ must be submitted to the manufacturer to
establish the incidence and benefit of laboratory
monitoring as well as other characteristics of these
2. Blood and platelet counts should be performed
before the initiation of CBZ therapy. Patients
with abnormalities on these tests and those taking
other myelotoxic drugs should be considered at
special risk. CBZ therapy in these groups should
be monitored closely or avoided.
3. Complete blood counts should be made every
two weeks for the first two months. If no abnormalities appear, counts should be obtained quarterly or with the appearance of symptoms or signs
of bone marrow suppression. Patients should be
instructed to contact their physician immediately
if petechiae, pallor, undue weakness, fever, or
infection occurs.
4. If leukopenia develops, it should be monitored at
two-week intervals seeking the expected return
to baseline. The drug dose should be reduced
if possible, and the presence of infection or
severe leukopenia (<3,000 WBC/mm3 or < 1,500
ne~ tr ophilslmm~
) for immediate cessation of
CBZ use 13, 41.
Hart and Easton: Laboratory Monitoring with Carbamazepine 3 11
1. Cereghino JJ, Brock JT, Van Meter JC, et al: Carbamazepine
for epilepsy. Neurology (Minneap) 24:401-410, 1974
2. Davis EH: Clinical trials of Tegretol in trigeminal deuralgia.
Headache 977-82, 1969
3. De Vries S1: Haematological aspects during treatment with
anticonvulsant drugs. Epilepsia 61-15, 1965
4. Gallagher BB: Neuropharmacology and the rrearnient of
epilepsy. In Vida JA (ed): Anticonvulsants. New York, Academic, 1977
5. Geigy Pharmaceutical Division: Package insert for Tegrerol
(carbamazepine). Ciba-Geigy Corporation, 1980
6. Gerber JG, Stiles G, Nies AS: Severe leukopenia secondary
to carbamazepine administration. South Med J 72:81-83,
7. Goldberg MA: Cosrs of anticonvulsant therapy. Ann Neurol
9:95, 1981
8. Hassan MN, Parsonage MJ: Experience in the long-term use
of carbamazepine (Tegrerol) in the treatment of epilepsy. In
Penry JK (ed): Epilepsy, The Eighth International Symposium. New York, Raven, 1977, pp 35-44
9. Hawson GAT, Bain BJ, Whyte I: Agranulocytosis after administration of carbamazepine. Med J Aust 1:82-83, 1980
10. Huf R, Schain RJ: Long-term experience with carbamazepine
(Tegretoi) in children with seizures. J Pediarr 97:310-312,
11. Killian JM: Tegretol in trigeminal neuralgia with special reference to hematopoietic side effects. Headache 958-63,
12. Killian JM, Fromm G H : Carbamazepine in the treatment of
neuralgia. IJse and side effects. Arch Neurol 19129-136,
13. Kosteljanetz M, Christiansen J, Dam AM, et al: Carbamazepine versus phenytoin: a controlled clinical trial. Arch
Neurol 36:22-24, 1979
14. Kurt H , Solomon G , Westerlain C, et al: Carbamazepine in
difficult to control epileptic out-patients. Acta Neurol Scand
60 [S~ppl]:27-32, 1975
15. Livingston S, Pauli LL, German W: Carbamazepine in
epilepsy. Nine year follow-up with special reference on untoward reactions. Dis Nerv Sys 35:103-107, 1974
16. Livingston S, Pauli LL, Pruce I: N o proven relationship of carbamazepine therapy to blood dyscrasias. Neurology (Minneap) 28:101-102, 1978
312 Annals of Neurology
Vol 11 No 3 March 1982
17. Livingston S, Villamater C, Sakata Y, er al: Use of carbamazepine in epilepsy. Results in 87 patients. JAMA
200:116-120, 1967
18. Meinardi H . Stoel LMK Side-effects of anti-epileptic drugs.
In Vinken PJ, Bruyn G W (eds): Handbook of Clinical
Neurology. New York, Elsevier, 1974, vol 15, chap 37
19. Monaco F, Riccio A, Benna P, et al: Further observations on
carbamazepine plasma levels in epileptic patients. Neurology
20. Murphy JE, Stewart RB, Springer PK: Carbamazepineinduced leukocytosis. Am J Hosp Pharm 37:550-552, 1980
2 1. Nicol CF: A four-year double blind study of Tegretol in facial
pain. Headache 9:54-57, 1969
22. Pearce J, Ron MA: Thrombocytopenia after carbamazepine.
Lancet 2:223, 1968
23. Pisciotta AV: Hematologic toxicity of carbamazepine. Adv
Neurol 11:355-368, 1975
24. Rodin EA, Rim CS, Rennick PM: The effects of carbamazepine on patients with psychomotor epilepsy: results of
a double-blind study. Epilepsia 15:547-561, 1974
25. Rutman JY: Effect of carbamazepine o n blood elements. Ann
Neurol 3:373, 1978
26. Schain RJ, Ward JW, Guthrie D: Carbamazepine as an anticonvulsant in children. Neurology (Minneap) 27:476-480,
27. Scheffner D, Schiefer 1: The treatment of epileptic children
with carbamazepine. Epilepsia 13:819-828, 1972
28. Schneider H : Long-term treatment in severe epilepsy (institutionalized patients): 11. Retrospective evaluation of carbamazepine. In Penry JK (ed): Epilepsy, The Eighth International Symposium. New York, Raven, 1977, pp 57-62
29. Singh AN, Saxena RM, Germain M: Anticonvulsive and
psychotrophic effects of carbamazepine in hospitalized
epileptic patients: a long-term study. In Penry JK (ed):
Epilepsy, The Eighth International Symposium. New York,
Raven, 1977, pp 47-56
30. Sturman R H , O’Brien FH: Non-surgical treatment of tic
douloureux with carbamazepine. Headache 9:88-91, 1969
31. Troupin A: The choice of anticonvulsanrs. In Laidlaw J,
Richens A (eds): Textbook of Epilepsy. New York,
Churchill/Livingstone, 1976
32. Troupin A, Ojemann LM, Halpern L, er al: Carbamazepine-a double-blind comparison with phenytoin. Neurology (Minneap) 27:511-519, 1977
Без категории
Размер файла
397 Кб
monitoring, carbamazepine, hematologic
Пожаловаться на содержимое документа