- Email: [email protected]
Neurophysiologic and intellectual evaluation of beta-thalassemia patients
Brain & Development 28 (2006) 14–18 www.elsevier.com/locate/braindev
Original article
Neurophysiologic and intellectual evaluation of beta-thalassemia patients Marina Economou, Dimitrios I. Zafeiriou*, Eleftherios Kontopoulos, Nikos Gompakis, Aphroditi Koussi, Vasilios Perifanis, Miranda Athanassiou-Metaxa 1st Department of Pediatrics, Aristotle University of Thessaloniki, and Thalassemia Unit, Hippokratio General Hospital, Thessaloniki, Egnatia St. 106, Thessaloniki 546 22, Greece Received 24 September 2004; received in revised form 16 March 2005; accepted 16 March 2005
Abstract In order to detect involvement of the central and peripheral nervous system in beta-thalassemic patients, 32 children and young adults (mean age 14.5G6.4 years) participated in a systematic neurophysiologic and intellectual prospective study. All patients were in a regular transfusion program, receiving subcutaneous desferrioxamine chelation and maintaining a mean serum ferritin level of 2,101.56G 986.32 ng/ml. Study patients underwent neurophysiologic evaluation consisting of brainstem auditory, visual and somatosensory evoked potential examination (BAEP, VEP, SEP) as well as motor and sensory nerve conduction velocity studies (MCV, SCV). Additionally, the verbal, performance and total IQ were assessed in patients under 16 years of age using the Weschler Intelligence Scale for Children (WISCIII). The incidence of abnormal BAEP, VEP, SEP and NCVs was 0, 3.12, 3.12 and 18.75%, respectively, findings comparative to or better than previously reported. On the contrary, the prevalence of abnormal total IQ score was considerably high (36.4%), not correlating, however, to any of the parameters assessed (age, sex, ferritin level, BAEP, VEP, SEP, NCV). Factors associated with chronic illness, rather than the disease per se, could play a potential role in the development of cognitive dysfunction in beta-thalassemia patients. q 2005 Elsevier B.V. All rights reserved. Keywords: Beta-thalassemia; Desferrioxamine; Neurotoxicity; Cognitive dysfunction; Neurophysiology; Evoked potentials; Nerve conduction velocity; Intelligence quotient
1. Introduction Beta-thalassemia major (BTM) is the most severe form of a group of inherited disorders of hemoglobin. Thalassemic patients require regular red cell transfusions in order to eliminate anemia complications and compensatory bone marrow expansion [1]. However, transfusions combined with excessive iron absorption lead to various organ iron deposition, principally affecting the heart, liver and endocrine glands and resulting in death, should patients be left untreated [2]. Profound changes in the management of BTM during the past decades have accomplished one of the most dramatic alterations in morbidity and mortality associated with a genetic disease. Deferoxamine mesylate (DFO), first introduced during the 1960s, gradually became the standard iron chelating therapy through out the world, * Corresponding author. Tel.: C30 2310 241845; fax: C30 23920 63186. E-mail address: [email protected] (D.I. Zafeiriou).
0387-7604/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2005.03.006
offering improved life quality and extended survival free of iron-induced complications in well-chelated patients [3]. During the last two decades several reports have demonstrated involvement of the nervous system in BTM patients, mainly concerning the ocular and auditory pathways [4–9]. In most reports, neurotoxicity has been attributed to high DFO doses, with complete or partial recovery observed following drug discontinuation [10,11]. Additional factors associated with impairment of the neural system in thalassemics include chronic hypoxia, bone marrow expansion and iron overload [12–15]. The object of the current study was to perform a systematic neurophysiologic and intellectual evaluation in order to detect possible involvement of the central and peripheral neural pathways in a group of neurologically asymptomatic BTM patients, consisting of both children and young adults.
2. Patients and methods For the purposes of the study 32 patients aged 6–32 years (mean age 14.5G6.39 years), 23 children and 9 young
M. Economou et al. / Brain & Development 28 (2006) 14–18
adults, were prospectively evaluated. All patients were in a regular transfusion program, in order to maintain a pretransfusion hemoglobin (Hb) level of 9–10 g/dl, and were receiving subcutaneous DFO chelation therapy on a mean daily dose of 30–50 mg/kg, 5–6 days weekly. Mean age at onset of blood transfusions was 0.9G0.1 years (range 0.7–1.0), mean age at onset of DFO therapy was 1.4G0.14 years (range 1.0–1.6 years), while, the mean duration between start of DFO therapy and the study was 13.1G6.4 years (range 4.5–30.6). Mean serum ferritin level in participating patients was 2,101.56G986.32 ng/ml, ranging from 800 to 5000 ng/ml. None of the patients had received prior treatment with drugs known to be neurotoxic, besides DFO, nor suffered from diabetes mellitus, chronic renal failure or chronic liver infection, i.e. conditions predisposing to nervous system complications. All participating patients underwent neurophysiologic evaluation consisting of brainstem auditory, visual and somatosensory evoked potential examination (BAEP, VEP, SEP), as well as motor (MCV) and sensory (SCV) nerve conduction velocity studies. Additionally, neuropsychologic examination evaluating the verbal, performance and total IQ domains was performed in patients under 16 years of age. Informed consent was obtained prior to patient participation in the study. None of the patients was ever tested neurophysiologically or intellectually prior to the beginning of the study. For the evoked potential (EP) study, the 10–20 International System of electrode placement was used. All neurophysiological studies were performed using the Nihon–Kohden Neuropack 8 machine in a soundproof and electrically shielded room at a room temperature of 24–25 8C, in order to keep a constant skin temperature. Silver chloride electrodes were used and impedance was kept below 5 kU. BAEP were recorded between ipsilateral ear and F2, in response to rarefaction click stimuli presented monaurally, at the rate of 10 Hz and at 70 dB above normal hearing level (bandpass 200–2000 Hz, speed duration 10 ms). Two averages each consisting of 1024 sweeps were performed and the latencies of waves I–V, as well as the inter-peak latencies I–III, III–V and I–V, were analysed. Pattern-reversal VEP were recorded over the inion (O2) referenced to F2, using pattern-reversal stimulation with a checkerboard pattern on a television screen and reversing at a rate of two times per second (bandpass 2–100 Hz, sweep speed 300 ms). Two averages each consisting of 128 trials were performed, with monocular stimulation of each eye, while all patients with refractive errors had their corrective lenses during the VEP study. The latency of the first positive peak (P100) as well as the inter-eye latency difference (ILD) were analysed. SEP were recorded over scalp central electrodes (C3 and C4) referenced to Fz, with stimulation of the median nerve at the wrist, with electrical impulses just above the threshold
15
of eliciting movement of the thumb (bandpass 2–100 Hz). Two averages each of 128 trials were performed and the latency to the Erb’s point (N9) and the cortical region (N20) was estimated. According to previous studies and taking into account that SEP of the median nerve is the more commonly used indicator of both a peripheral as well as central involvement of the somatosensory pathways, only median nerve SEP was performed in the current study. MCV of the median and ulnar nerve was performed using supramaximal stimulation of the nerves at the wrist and elbow, with recording over the abductor pollicis brevis and abductor digiti minimi, respectively. The MCV of the peroneal and posterior tibial nerve was obtained by supramaximal stimulation of the nerves at the ankle and knee, with recording over the extensor digitorum brevis and adductor halluci, respectively. The SCV of the median and ulnar nerve was antidromically recorded by stimulating the nerves at the wrist and elbow over the 1st and 5th digit of the upper extremity, respectively, using ring finger electrodes. Sural nerve conduction study was not performed due to the technical difficulties that poses its recording, especially in small children, a fact that could lead to unreliable results. Intellectual evaluation was performed in all subjects aged between 6 and 16 years, using the Weschler Intelligence Scale for Children-3rd edition (Greek standardization) [16]. The evaluation was performed by a qualified clinical child psychologist, within a time frame G2 weeks apart from the neurophysiological study. The examiner was aware that the subjects were BTM patients and were participating in a study, however, she was unaware of other study results. Patients older than 16 years were not tested due to non-availability of a clinical psychologist specialized in adults in our institution during the time of the study. Control data were obtained for 40 normal subjects (own laboratory data), of a similar age range with regards to BAEP, VEP and NCV and of similar height range with regards to SEP. According to previous literature [17], 2 age groups were used with regards to NCV evaluation (age range 6–16 and 20–32 years, respectively). Abnormalities in EP and NCV studies were defined as those with latencies two or more standard deviations from the norm in a given patient -compared to controls-, without necessarily reflecting abnormalities in the whole group of BTM patients tested. Statistical calculations were conducted using one-way analyses of variance (independent t-test and one-way ANOVA), as well as descriptive statistics.
3. Results Study results regarding neurophysiologic and intellectual evaluation are summarized in Table 1. It is important to note that the involvement of the nervous system was subclinical in all cases, i.e. none of the patients demonstrating abnormal
16
M. Economou et al. / Brain & Development 28 (2006) 14–18
Table 1 Results of neurophysiologic and neuropsychologic evaluation Parameters
Patients Mean value
VEP: P100 Left Right ILD BAEP: I Left Right BAEP: I–V Left Right BAEP: I–III Left Right BAEP: III–V Left Right SEP: N9 Left Right SEP: N20 Left Right MCV-median Children Adults MCV-ulnar Children Adults MCV-peroneal Children Adults MCV-pos. tibial Children Adults SCV-median Children Adults SCV-ulnar Children Adults IQ V IQ P IQ T
Controls SD
Mean value
SD
106.90
9.50
Statistical significance
105.83 106.1 2.39
4.78 5.38 2.07
1.74 1.78
0.12 0.17
3.98 4.00
0.20 0.26
2.15 2.16
0.14 0.23
1.83 1.84
0.16 0.15
9.54 9.45
2.01 1.11
17.51 17.20
2.48 2.88
50.30 52.00
2.67 3.51
57.20 57.80
3.70 2.60
P!0.001 NS
48.10 51.50
4.33 3.52
58.10 61.90
9.60 5.10
NS PZ0.003
43.20 39.80
4.27 4.43
57.00 49.90
4.50 5.90
P!0.001 PZ0.002
40.10 42.70
6.28 5.11
48.20 49.80
2.70 4.40
PZ0.026 NS
39.60 42.30
4.33 3.47
43.70 47.50
3.40 4.60
PZ0.013 NS
46.41 47.43 94.55 88.27 90.55
8.22 6.40 19.99 14.62 18.59
43.90 55.40 – – –
3.90 3.60 – – –
NS PZ0.006 – – –
3.20 1.75
1.80 0.20
NS NS NS NS NS
4.10
0.30 PZ0.002 PZ0.032
2.1
0.15 NS NS
1.9
0.18 PZ0.019 PZ0.031
9.60
0.70 NS NS
18.50
1.00 NS NS
VEP, visual; BAEP, brainstem auditory; SEP, somatosensory evoked potentials; MCV, motor nerve conduction velocity; SCV, sensory nerve conduction velocity; V, verbal; P, performance; T, total; NS, nonsignificant.
neurophysiological results presented relevant signs or symptoms. With regard to BAEP, none of the patients demonstrated abnormal findings and mean wave I, as well as inter-peak wave I–III latencies, did not statistically differ significantly compared to controls. In the contrary, mean inter-peak latencies I–V and III–V were bilaterally reduced compared to control group values.
The incidence of VEP involvement was very low (3.12%), with only one patient demonstrating increased inter-eye latency difference of 10.2 ms (upper normal limit 6.8 ms according to controls). Mean P100 wave latencies were bilaterally normal compared to mean control values. Regarding SEP, bilateral prolongation of N20 wave latency was observed in one patient (3.12%). Mean SEP values were within normal limits compared to controls. With regard to NCV, 6 patients in total presented abnormal findings (18.75%). Two of them solely demonstrated pathologic MCV, while all 6 showed abnormal SCV. There was no statistically significant relationship between NCV studies and compared parameters, i.e. sex, age, weight, height, serum ferritin, BAEP, VEP and SEP, while there was a statistically significant slowing in most nerves examined compared to controls (Table 1). The prevalence of pathological total IQ score was considerably high (36.36%), 8 of the 22 children assessed presenting IQ scores under 85 (Table 2). Interestingly enough, in 3 out of the 8 patients mentioned IQ score was under 70, indicating mild mental retardation. None of the 8 patients presented additional factors predisposing to decreased cognitive function (such as possible syndromes, positive family history etc.). Abnormal IQ score did not correlate statistically to parameters assessed (sex, age, ferritin level, BAEP, VEP, SEP, NCV).
4. Discussion Neural pathway involvement in BTM patients has been a study object for many years, being attributed at times to various factors such as hypoxia, hemosiderosis, diabetes mellitus and DFO neurotoxicity [14,18,19]. Different reports, however, have mainly focused on visual and auditory neuropathy, overlooking possible clinical or subclinical involvement of other neural pathways. To our knowledge, this is the first systematic neurophysiologic and intellectual study investigating possible complications of the central and peripheral nervous system in BTM patients. The incidence of abnormal BAEP and VEP worldwide is found to be considerably high compared to the present study, attributed in most cases to DFO neurotoxicity [9,20] and less often to hemosiderosis [18]. With regard to SEP, relevant literature is limited and mostly associated to diabetes [14], a complication usually found in poorly chelated BTM patients. It should be noted that in an older study conducted in the same Thalassemia Unit, the incidence of abnormal EP was equally high to that reported in other centers, with pathologic BAEP, VEP and SEP detected in 25, 15 and 7.5% of the patients evaluated, respectively, [21]. Based on those findings and in accordance to reported literature [10], the therapeutic index -defined as the ratio of mean daily DFO dosage divided by the serum ferritin level and considered to be a risk factor when exceeding 0.025-, was
M. Economou et al. / Brain & Development 28 (2006) 14–18
17
Table 2 WISC analytical scores and subscales of NZ22 patients with BTM aged 6–16 years Pat.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
WISC subscales
IQ
A
B
C
D
E
F
G
H
I
J
V
P
T
18 17 9 24 19 4 22 21 18 20 20 23 20 16 17 10 18 10 12 20 19 12
21 17 9 28 22 6 25 12 12 15 20 23 16 10 14 8 8 5 14 15 19 16
47 50 19 68 32 34 54 39 36 40 56 45 68 48 38 40 45 31 32 40 67 45
18 10 11 26 22 7 22 9 9 12 17 22 17 10 11 7 7 7 11 13 19 16
19 13 4 47 34 12 28 21 12 39 22 30 23 21 15 13 17 4 26 3 32 11
19 17 12 22 16 10 21 16 16 17 20 22 26 15 15 11 12 14 15 14 17 15
41 45 5 56 40 6 46 26 35 37 40 54 53 31 15 5 28 10 20 23 47 27
39 26 14 49 41 7 44 25 14 27 36 40 27 23 26 17 17 14 29 16 45 25
27 29 6 32 27 6 31 24 21 25 27 29 24 20 13 7 22 11 11 25 33 15
23 16 6 30 25 9 28 19 19 23 23 27 25 20 21 10 18 9 18 18 22 20
111 88 113 121 109 82 123 83 65 110 86 124 104 80 99 100 58 96 104 53 83 88
93 97 77 113 92 79 100 87 66 119 82 107 102 86 83 89 76 85 90 59 90 70
103 92 96 120 101 78 113 83 61 117 82 118 103 81 90 94 62 89 97 50 85 77
A, Picture completion; B, Information; C, Coding; D, Similarity; E, Picture arrangement; F, Arithmetic; G, Block design; H, Vocabulary; I, Object assembly; J, Comprehension; IQ, intelligence quotient; V, verbal; P, performance; T, total; Numbers in bold: abnormal total IQ scores.
applied thereafter in all unit patients. According to the index mentioned, DFO neurotoxicity is not only associated with high drug doses, but can also result from regular dosage in patients with relatively low iron load. The above mentioned encouraging EP results could, therefore, be attributed to proper DFO dosing and regular iron burden monitoring in participating patients, combined with a lack of other predisposing factors, such as diabetes mellitus or evident hypoxia. Individual cases described above, involving abnormal VEP and SEP, cannot be used in order to conduct safe conclusions. One can only hypothesize that hemosiderosis resulting from long term poor chelation compliance or DFO toxicity associated with a low iron content could be the cause of abnormal findings in the relevant cases. One, however, should bear in mind that DFO toxicity is probably related to individual susceptibility and, therefore, cannot always be foreseeable or easy to prevent [4]. The results of the NCV studies point out to a subclinical peripheral polyneuropathy, since none of the patients presented with signs or symptoms associated with clinically evident peripheral neuropathy (reduced or absent tendon reflexes, muscle weakness and/or atrophy, sensory symptoms etc.). Literature concerning NCV studies in BTM patients remains limited. Sporadic studies show results comparable to those herein reported, with regards to abnormal MCV and/or SCV values, attributed either to chronic hypoxia and older age [22], or to hemosiderosis, whether by means of pancreas involvement or not [14]. In the present study, mild
NCV slowing could not be associated with any of the risk factors specified; pathologic values involved well transfused patients maintaining satisfactory hemoglobin levels, not suffering from diabetes mellitus and, all but one, showing good compliance to chelation therapy with a mean serum ferritin level of %2,500 ng/ml. Study results concerning intellectual evaluation are, in our opinion, of particular interest; abnormal IQ scores (!85) were present in an unexpectedly high percentage of evaluated patients (36.36%) and in three cases indicated mild mental retardation (!70). These abnormal IQ scores were not associated with or statistically correlated to factors predisposing to cognitive dysfunction. Relevant literature is limited, not to date, and presenting conflicting results [23–25]. The sole recent report studying multiple cognitive domains, including, however, only BTM patients aged over 16 years, argues for a potential role of hemosiderosis in cognitive functioning [26], a parameter not correlating to the present study results. Cognitive dysfunction in a BTM patient group well transfused and adequately chelated, as the one studied in the current report, could possibly—even if partly—be attributed to factors associated to severe chronic illness, rather than the disease per se. Such factors include regular school absence due to transfusions and frequent hospitalizations, physical and social restrictions resulting from the disease or its treatment, abnormal mental state due to the awareness of being chronically ill or the threats imposed by the disease and last, but not least, the overly protective family attitude
18
M. Economou et al. / Brain & Development 28 (2006) 14–18
that leads to restricted initiative and psychosocial development. Conclusively, BTM patients demonstrate subclinical involvement of the central and peripheral neural pathways, therefore, regular neurophysiologic and especially intellectual monitoring (WISC-III) is imperative in order to detect relevant abnormalities and apply appropriate management. We recommend that the WISC-III test should be applied as early as possible (from 6 years of age) in order to detect as early as possible any kind of intellectual dysfunction in young patients with beta-thalassemia.
References [1] Fosburg M, Nathan D. Treatment of Cooley’s anemia. Blood 1990;76: 435–44. [2] Modell B, Khan M, Darlison M. Survival in beta-thalassemia major in the UK: data from the UK Thalassemia Register. Lancet 2000;355: 2051–2. [3] Olivieri N, Brittengham G. Iron chelation therapy and the treatment of thalassemia. Blood 1997;89:739–61. [4] Ambrosetti U, Donde E, Piatti G, Capellini M. Audiological evaluation in adult beta-thalassemia major patients under regular chelation treatment. Pharmacol Res 2000;42:485–7. [5] Chiodo A, Alberti P, Sher G, Francombe W, Tayler B. Desferrioxamine ototoxicity in an adult transfusion dependent population. J Otolaryngol 1997;26:116–22. [6] Borgna-Pignati C, De Stefano P, Broglia A. Visual loss in patient on high-dose subcutaneous desferrioxamine. Lancet 1984;1:681. [7] De Virgiliis S, Congia M, Turco M, Frau F, Dessi C, Argiolou F, et al. Depletion of trace elements and acute ocular toxicity induced by desferrioxamine in patients with thalassemia. Arch Dis Child 1988; 63:250–5. [8] Porter J. A risk-benefit assessment of iron-chelation therapy. Drug Saf 1997;17:407–21. [9] Olivieri N, Buncic J, Chew E, Gallant T, Harrison R, Keenan N, et al. Visual and auditory neurotoxicity in patients receiving subcutaneous desferrioxamine infusions. New Engl J Med 1986;314:869–73. [10] Porter J, Jawson M, Huehns East C, Hazell J. Desferrioxamine ototoxicity: evaluation of risk factors in thalassaemic patients and guidelines for safe dosage. Br J Haematol 1989;73:403–9. [11] Wonke B, Hoffbrand A, Aldouri M, Wickens D, Flynn D, Stearns M, et al. Reversal of desferrioxamine induced auditory neurotoxicity during treatment with Ca-DTPA. Arch Dis Child 1989;64:77–82.
[12] Aarabi B, Haghshenas M, Rakeii V. Visual failure caused by suprasellar extramedullary hematopoiesis in beta thalassemia: case report. Neurosurgery 1998;42:922–5. [13] Cohen A, Martin M, Mizanin J, Konkle D, Schwartz E. Vision and hearing during deferoxamine therapy. J Pediatr 1990;117: 326–30. [14] Wong V, Li A, Lee A. Neurophysiologic study of beta-thalassemia patients. J Child Neurol 1993;8:330–5. [15] Incorpora G, Di Gregorio F, Romeo M, Pavone P, Trifiletti R, Parano E. Focal neurologic deficits in children with beta-thalassemia major. Neuropediatrics 1999;30:45–8. [16] Georgas D, Paraskevopoulos I, Bezebegis I, Giannitsas N. Weschler Intelligence Scale for Children-III-Manual. Greek ed. Athens: Hellenic Letters; 1997. [17] Cai F, Zhang J. Study of nerve conduction and late responses in normal Chinese infants, children and adults. J Child Neurol 1997;12: 13–18. [18] Gelmi C, Borgna-Pignatti C, Franchin S, Tacchini M, Trimarchi F. Electroretinographic and visual evoked potential abnormalities in patients with beta-thalassemia major. Ophthalmologica 1988;196: 29–34. [19] Bentur Y, Koren G, Tesoro A, Carley H, Olivieri N, Freedman M. Comparison of deferoxamine pharmacokinetics between asymptomatic thalassemic children and those exhibiting severe neurotoxicity. Clin Pharmacol Ther 1990;47:478–82. [20] Amabile G, Stefano E, Bianco I, Aliquo M, Giunti P, Pierelli F. Electrophysiological study in patients with beta-thalassemia major. Acta Neurol 1987;87:181–90. [21] Zafeiriou D, Kousi A, Tsantali C, Kontopoulos E, AugoustidouSavvopoulou P, Tsoubaris P, et al. Neurophysiologic evaluation of long-term desferrioxamine therapy in beta-thalassemia patients. Pediatr Neurol 1998;18:420–4. [22] Papanastasiou D, Papanicolaou D, Magiakou A, Beratis N, Tzebelikos E, Papapetropoulos T. Peripheral neuropathy in patients with beta-thalassemia. J Neurol Neurosurg Psychiatry 1991;54: 997–1000. [23] Davidson C, Weschler I. Erythroblastic (Cooley’s) anemia and neurologic complications. Am J Dis Child 1939;58:362–5. [24] Matoth Y, Shamir Z, Freunollich E. Thalassemia in Jews from Kurdistan. Blood 1955;10:176–8. [25] Logothetis J, Haritos M, Constantoulakis M, Economidou J, Augoustaki O, Lowenson R. Intelligence and behavioral patterns in patients with Cooley’s anemia; a study based on 138 consecutive cases. Pediatrics 1971;48:740–4. [26] Monastero R, Monastero G, Ciaccio C, Padovani A, Camadra R. Cognitive deficits in beta-thalassemia major. Acta Neurol Scand 2000;102:162–8.
Original article
Neurophysiologic and intellectual evaluation of beta-thalassemia patients Marina Economou, Dimitrios I. Zafeiriou*, Eleftherios Kontopoulos, Nikos Gompakis, Aphroditi Koussi, Vasilios Perifanis, Miranda Athanassiou-Metaxa 1st Department of Pediatrics, Aristotle University of Thessaloniki, and Thalassemia Unit, Hippokratio General Hospital, Thessaloniki, Egnatia St. 106, Thessaloniki 546 22, Greece Received 24 September 2004; received in revised form 16 March 2005; accepted 16 March 2005
Abstract In order to detect involvement of the central and peripheral nervous system in beta-thalassemic patients, 32 children and young adults (mean age 14.5G6.4 years) participated in a systematic neurophysiologic and intellectual prospective study. All patients were in a regular transfusion program, receiving subcutaneous desferrioxamine chelation and maintaining a mean serum ferritin level of 2,101.56G 986.32 ng/ml. Study patients underwent neurophysiologic evaluation consisting of brainstem auditory, visual and somatosensory evoked potential examination (BAEP, VEP, SEP) as well as motor and sensory nerve conduction velocity studies (MCV, SCV). Additionally, the verbal, performance and total IQ were assessed in patients under 16 years of age using the Weschler Intelligence Scale for Children (WISCIII). The incidence of abnormal BAEP, VEP, SEP and NCVs was 0, 3.12, 3.12 and 18.75%, respectively, findings comparative to or better than previously reported. On the contrary, the prevalence of abnormal total IQ score was considerably high (36.4%), not correlating, however, to any of the parameters assessed (age, sex, ferritin level, BAEP, VEP, SEP, NCV). Factors associated with chronic illness, rather than the disease per se, could play a potential role in the development of cognitive dysfunction in beta-thalassemia patients. q 2005 Elsevier B.V. All rights reserved. Keywords: Beta-thalassemia; Desferrioxamine; Neurotoxicity; Cognitive dysfunction; Neurophysiology; Evoked potentials; Nerve conduction velocity; Intelligence quotient
1. Introduction Beta-thalassemia major (BTM) is the most severe form of a group of inherited disorders of hemoglobin. Thalassemic patients require regular red cell transfusions in order to eliminate anemia complications and compensatory bone marrow expansion [1]. However, transfusions combined with excessive iron absorption lead to various organ iron deposition, principally affecting the heart, liver and endocrine glands and resulting in death, should patients be left untreated [2]. Profound changes in the management of BTM during the past decades have accomplished one of the most dramatic alterations in morbidity and mortality associated with a genetic disease. Deferoxamine mesylate (DFO), first introduced during the 1960s, gradually became the standard iron chelating therapy through out the world, * Corresponding author. Tel.: C30 2310 241845; fax: C30 23920 63186. E-mail address: [email protected] (D.I. Zafeiriou).
0387-7604/$ - see front matter q 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2005.03.006
offering improved life quality and extended survival free of iron-induced complications in well-chelated patients [3]. During the last two decades several reports have demonstrated involvement of the nervous system in BTM patients, mainly concerning the ocular and auditory pathways [4–9]. In most reports, neurotoxicity has been attributed to high DFO doses, with complete or partial recovery observed following drug discontinuation [10,11]. Additional factors associated with impairment of the neural system in thalassemics include chronic hypoxia, bone marrow expansion and iron overload [12–15]. The object of the current study was to perform a systematic neurophysiologic and intellectual evaluation in order to detect possible involvement of the central and peripheral neural pathways in a group of neurologically asymptomatic BTM patients, consisting of both children and young adults.
2. Patients and methods For the purposes of the study 32 patients aged 6–32 years (mean age 14.5G6.39 years), 23 children and 9 young
M. Economou et al. / Brain & Development 28 (2006) 14–18
adults, were prospectively evaluated. All patients were in a regular transfusion program, in order to maintain a pretransfusion hemoglobin (Hb) level of 9–10 g/dl, and were receiving subcutaneous DFO chelation therapy on a mean daily dose of 30–50 mg/kg, 5–6 days weekly. Mean age at onset of blood transfusions was 0.9G0.1 years (range 0.7–1.0), mean age at onset of DFO therapy was 1.4G0.14 years (range 1.0–1.6 years), while, the mean duration between start of DFO therapy and the study was 13.1G6.4 years (range 4.5–30.6). Mean serum ferritin level in participating patients was 2,101.56G986.32 ng/ml, ranging from 800 to 5000 ng/ml. None of the patients had received prior treatment with drugs known to be neurotoxic, besides DFO, nor suffered from diabetes mellitus, chronic renal failure or chronic liver infection, i.e. conditions predisposing to nervous system complications. All participating patients underwent neurophysiologic evaluation consisting of brainstem auditory, visual and somatosensory evoked potential examination (BAEP, VEP, SEP), as well as motor (MCV) and sensory (SCV) nerve conduction velocity studies. Additionally, neuropsychologic examination evaluating the verbal, performance and total IQ domains was performed in patients under 16 years of age. Informed consent was obtained prior to patient participation in the study. None of the patients was ever tested neurophysiologically or intellectually prior to the beginning of the study. For the evoked potential (EP) study, the 10–20 International System of electrode placement was used. All neurophysiological studies were performed using the Nihon–Kohden Neuropack 8 machine in a soundproof and electrically shielded room at a room temperature of 24–25 8C, in order to keep a constant skin temperature. Silver chloride electrodes were used and impedance was kept below 5 kU. BAEP were recorded between ipsilateral ear and F2, in response to rarefaction click stimuli presented monaurally, at the rate of 10 Hz and at 70 dB above normal hearing level (bandpass 200–2000 Hz, speed duration 10 ms). Two averages each consisting of 1024 sweeps were performed and the latencies of waves I–V, as well as the inter-peak latencies I–III, III–V and I–V, were analysed. Pattern-reversal VEP were recorded over the inion (O2) referenced to F2, using pattern-reversal stimulation with a checkerboard pattern on a television screen and reversing at a rate of two times per second (bandpass 2–100 Hz, sweep speed 300 ms). Two averages each consisting of 128 trials were performed, with monocular stimulation of each eye, while all patients with refractive errors had their corrective lenses during the VEP study. The latency of the first positive peak (P100) as well as the inter-eye latency difference (ILD) were analysed. SEP were recorded over scalp central electrodes (C3 and C4) referenced to Fz, with stimulation of the median nerve at the wrist, with electrical impulses just above the threshold
15
of eliciting movement of the thumb (bandpass 2–100 Hz). Two averages each of 128 trials were performed and the latency to the Erb’s point (N9) and the cortical region (N20) was estimated. According to previous studies and taking into account that SEP of the median nerve is the more commonly used indicator of both a peripheral as well as central involvement of the somatosensory pathways, only median nerve SEP was performed in the current study. MCV of the median and ulnar nerve was performed using supramaximal stimulation of the nerves at the wrist and elbow, with recording over the abductor pollicis brevis and abductor digiti minimi, respectively. The MCV of the peroneal and posterior tibial nerve was obtained by supramaximal stimulation of the nerves at the ankle and knee, with recording over the extensor digitorum brevis and adductor halluci, respectively. The SCV of the median and ulnar nerve was antidromically recorded by stimulating the nerves at the wrist and elbow over the 1st and 5th digit of the upper extremity, respectively, using ring finger electrodes. Sural nerve conduction study was not performed due to the technical difficulties that poses its recording, especially in small children, a fact that could lead to unreliable results. Intellectual evaluation was performed in all subjects aged between 6 and 16 years, using the Weschler Intelligence Scale for Children-3rd edition (Greek standardization) [16]. The evaluation was performed by a qualified clinical child psychologist, within a time frame G2 weeks apart from the neurophysiological study. The examiner was aware that the subjects were BTM patients and were participating in a study, however, she was unaware of other study results. Patients older than 16 years were not tested due to non-availability of a clinical psychologist specialized in adults in our institution during the time of the study. Control data were obtained for 40 normal subjects (own laboratory data), of a similar age range with regards to BAEP, VEP and NCV and of similar height range with regards to SEP. According to previous literature [17], 2 age groups were used with regards to NCV evaluation (age range 6–16 and 20–32 years, respectively). Abnormalities in EP and NCV studies were defined as those with latencies two or more standard deviations from the norm in a given patient -compared to controls-, without necessarily reflecting abnormalities in the whole group of BTM patients tested. Statistical calculations were conducted using one-way analyses of variance (independent t-test and one-way ANOVA), as well as descriptive statistics.
3. Results Study results regarding neurophysiologic and intellectual evaluation are summarized in Table 1. It is important to note that the involvement of the nervous system was subclinical in all cases, i.e. none of the patients demonstrating abnormal
16
M. Economou et al. / Brain & Development 28 (2006) 14–18
Table 1 Results of neurophysiologic and neuropsychologic evaluation Parameters
Patients Mean value
VEP: P100 Left Right ILD BAEP: I Left Right BAEP: I–V Left Right BAEP: I–III Left Right BAEP: III–V Left Right SEP: N9 Left Right SEP: N20 Left Right MCV-median Children Adults MCV-ulnar Children Adults MCV-peroneal Children Adults MCV-pos. tibial Children Adults SCV-median Children Adults SCV-ulnar Children Adults IQ V IQ P IQ T
Controls SD
Mean value
SD
106.90
9.50
Statistical significance
105.83 106.1 2.39
4.78 5.38 2.07
1.74 1.78
0.12 0.17
3.98 4.00
0.20 0.26
2.15 2.16
0.14 0.23
1.83 1.84
0.16 0.15
9.54 9.45
2.01 1.11
17.51 17.20
2.48 2.88
50.30 52.00
2.67 3.51
57.20 57.80
3.70 2.60
P!0.001 NS
48.10 51.50
4.33 3.52
58.10 61.90
9.60 5.10
NS PZ0.003
43.20 39.80
4.27 4.43
57.00 49.90
4.50 5.90
P!0.001 PZ0.002
40.10 42.70
6.28 5.11
48.20 49.80
2.70 4.40
PZ0.026 NS
39.60 42.30
4.33 3.47
43.70 47.50
3.40 4.60
PZ0.013 NS
46.41 47.43 94.55 88.27 90.55
8.22 6.40 19.99 14.62 18.59
43.90 55.40 – – –
3.90 3.60 – – –
NS PZ0.006 – – –
3.20 1.75
1.80 0.20
NS NS NS NS NS
4.10
0.30 PZ0.002 PZ0.032
2.1
0.15 NS NS
1.9
0.18 PZ0.019 PZ0.031
9.60
0.70 NS NS
18.50
1.00 NS NS
VEP, visual; BAEP, brainstem auditory; SEP, somatosensory evoked potentials; MCV, motor nerve conduction velocity; SCV, sensory nerve conduction velocity; V, verbal; P, performance; T, total; NS, nonsignificant.
neurophysiological results presented relevant signs or symptoms. With regard to BAEP, none of the patients demonstrated abnormal findings and mean wave I, as well as inter-peak wave I–III latencies, did not statistically differ significantly compared to controls. In the contrary, mean inter-peak latencies I–V and III–V were bilaterally reduced compared to control group values.
The incidence of VEP involvement was very low (3.12%), with only one patient demonstrating increased inter-eye latency difference of 10.2 ms (upper normal limit 6.8 ms according to controls). Mean P100 wave latencies were bilaterally normal compared to mean control values. Regarding SEP, bilateral prolongation of N20 wave latency was observed in one patient (3.12%). Mean SEP values were within normal limits compared to controls. With regard to NCV, 6 patients in total presented abnormal findings (18.75%). Two of them solely demonstrated pathologic MCV, while all 6 showed abnormal SCV. There was no statistically significant relationship between NCV studies and compared parameters, i.e. sex, age, weight, height, serum ferritin, BAEP, VEP and SEP, while there was a statistically significant slowing in most nerves examined compared to controls (Table 1). The prevalence of pathological total IQ score was considerably high (36.36%), 8 of the 22 children assessed presenting IQ scores under 85 (Table 2). Interestingly enough, in 3 out of the 8 patients mentioned IQ score was under 70, indicating mild mental retardation. None of the 8 patients presented additional factors predisposing to decreased cognitive function (such as possible syndromes, positive family history etc.). Abnormal IQ score did not correlate statistically to parameters assessed (sex, age, ferritin level, BAEP, VEP, SEP, NCV).
4. Discussion Neural pathway involvement in BTM patients has been a study object for many years, being attributed at times to various factors such as hypoxia, hemosiderosis, diabetes mellitus and DFO neurotoxicity [14,18,19]. Different reports, however, have mainly focused on visual and auditory neuropathy, overlooking possible clinical or subclinical involvement of other neural pathways. To our knowledge, this is the first systematic neurophysiologic and intellectual study investigating possible complications of the central and peripheral nervous system in BTM patients. The incidence of abnormal BAEP and VEP worldwide is found to be considerably high compared to the present study, attributed in most cases to DFO neurotoxicity [9,20] and less often to hemosiderosis [18]. With regard to SEP, relevant literature is limited and mostly associated to diabetes [14], a complication usually found in poorly chelated BTM patients. It should be noted that in an older study conducted in the same Thalassemia Unit, the incidence of abnormal EP was equally high to that reported in other centers, with pathologic BAEP, VEP and SEP detected in 25, 15 and 7.5% of the patients evaluated, respectively, [21]. Based on those findings and in accordance to reported literature [10], the therapeutic index -defined as the ratio of mean daily DFO dosage divided by the serum ferritin level and considered to be a risk factor when exceeding 0.025-, was
M. Economou et al. / Brain & Development 28 (2006) 14–18
17
Table 2 WISC analytical scores and subscales of NZ22 patients with BTM aged 6–16 years Pat.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
WISC subscales
IQ
A
B
C
D
E
F
G
H
I
J
V
P
T
18 17 9 24 19 4 22 21 18 20 20 23 20 16 17 10 18 10 12 20 19 12
21 17 9 28 22 6 25 12 12 15 20 23 16 10 14 8 8 5 14 15 19 16
47 50 19 68 32 34 54 39 36 40 56 45 68 48 38 40 45 31 32 40 67 45
18 10 11 26 22 7 22 9 9 12 17 22 17 10 11 7 7 7 11 13 19 16
19 13 4 47 34 12 28 21 12 39 22 30 23 21 15 13 17 4 26 3 32 11
19 17 12 22 16 10 21 16 16 17 20 22 26 15 15 11 12 14 15 14 17 15
41 45 5 56 40 6 46 26 35 37 40 54 53 31 15 5 28 10 20 23 47 27
39 26 14 49 41 7 44 25 14 27 36 40 27 23 26 17 17 14 29 16 45 25
27 29 6 32 27 6 31 24 21 25 27 29 24 20 13 7 22 11 11 25 33 15
23 16 6 30 25 9 28 19 19 23 23 27 25 20 21 10 18 9 18 18 22 20
111 88 113 121 109 82 123 83 65 110 86 124 104 80 99 100 58 96 104 53 83 88
93 97 77 113 92 79 100 87 66 119 82 107 102 86 83 89 76 85 90 59 90 70
103 92 96 120 101 78 113 83 61 117 82 118 103 81 90 94 62 89 97 50 85 77
A, Picture completion; B, Information; C, Coding; D, Similarity; E, Picture arrangement; F, Arithmetic; G, Block design; H, Vocabulary; I, Object assembly; J, Comprehension; IQ, intelligence quotient; V, verbal; P, performance; T, total; Numbers in bold: abnormal total IQ scores.
applied thereafter in all unit patients. According to the index mentioned, DFO neurotoxicity is not only associated with high drug doses, but can also result from regular dosage in patients with relatively low iron load. The above mentioned encouraging EP results could, therefore, be attributed to proper DFO dosing and regular iron burden monitoring in participating patients, combined with a lack of other predisposing factors, such as diabetes mellitus or evident hypoxia. Individual cases described above, involving abnormal VEP and SEP, cannot be used in order to conduct safe conclusions. One can only hypothesize that hemosiderosis resulting from long term poor chelation compliance or DFO toxicity associated with a low iron content could be the cause of abnormal findings in the relevant cases. One, however, should bear in mind that DFO toxicity is probably related to individual susceptibility and, therefore, cannot always be foreseeable or easy to prevent [4]. The results of the NCV studies point out to a subclinical peripheral polyneuropathy, since none of the patients presented with signs or symptoms associated with clinically evident peripheral neuropathy (reduced or absent tendon reflexes, muscle weakness and/or atrophy, sensory symptoms etc.). Literature concerning NCV studies in BTM patients remains limited. Sporadic studies show results comparable to those herein reported, with regards to abnormal MCV and/or SCV values, attributed either to chronic hypoxia and older age [22], or to hemosiderosis, whether by means of pancreas involvement or not [14]. In the present study, mild
NCV slowing could not be associated with any of the risk factors specified; pathologic values involved well transfused patients maintaining satisfactory hemoglobin levels, not suffering from diabetes mellitus and, all but one, showing good compliance to chelation therapy with a mean serum ferritin level of %2,500 ng/ml. Study results concerning intellectual evaluation are, in our opinion, of particular interest; abnormal IQ scores (!85) were present in an unexpectedly high percentage of evaluated patients (36.36%) and in three cases indicated mild mental retardation (!70). These abnormal IQ scores were not associated with or statistically correlated to factors predisposing to cognitive dysfunction. Relevant literature is limited, not to date, and presenting conflicting results [23–25]. The sole recent report studying multiple cognitive domains, including, however, only BTM patients aged over 16 years, argues for a potential role of hemosiderosis in cognitive functioning [26], a parameter not correlating to the present study results. Cognitive dysfunction in a BTM patient group well transfused and adequately chelated, as the one studied in the current report, could possibly—even if partly—be attributed to factors associated to severe chronic illness, rather than the disease per se. Such factors include regular school absence due to transfusions and frequent hospitalizations, physical and social restrictions resulting from the disease or its treatment, abnormal mental state due to the awareness of being chronically ill or the threats imposed by the disease and last, but not least, the overly protective family attitude
18
M. Economou et al. / Brain & Development 28 (2006) 14–18
that leads to restricted initiative and psychosocial development. Conclusively, BTM patients demonstrate subclinical involvement of the central and peripheral neural pathways, therefore, regular neurophysiologic and especially intellectual monitoring (WISC-III) is imperative in order to detect relevant abnormalities and apply appropriate management. We recommend that the WISC-III test should be applied as early as possible (from 6 years of age) in order to detect as early as possible any kind of intellectual dysfunction in young patients with beta-thalassemia.
References [1] Fosburg M, Nathan D. Treatment of Cooley’s anemia. Blood 1990;76: 435–44. [2] Modell B, Khan M, Darlison M. Survival in beta-thalassemia major in the UK: data from the UK Thalassemia Register. Lancet 2000;355: 2051–2. [3] Olivieri N, Brittengham G. Iron chelation therapy and the treatment of thalassemia. Blood 1997;89:739–61. [4] Ambrosetti U, Donde E, Piatti G, Capellini M. Audiological evaluation in adult beta-thalassemia major patients under regular chelation treatment. Pharmacol Res 2000;42:485–7. [5] Chiodo A, Alberti P, Sher G, Francombe W, Tayler B. Desferrioxamine ototoxicity in an adult transfusion dependent population. J Otolaryngol 1997;26:116–22. [6] Borgna-Pignati C, De Stefano P, Broglia A. Visual loss in patient on high-dose subcutaneous desferrioxamine. Lancet 1984;1:681. [7] De Virgiliis S, Congia M, Turco M, Frau F, Dessi C, Argiolou F, et al. Depletion of trace elements and acute ocular toxicity induced by desferrioxamine in patients with thalassemia. Arch Dis Child 1988; 63:250–5. [8] Porter J. A risk-benefit assessment of iron-chelation therapy. Drug Saf 1997;17:407–21. [9] Olivieri N, Buncic J, Chew E, Gallant T, Harrison R, Keenan N, et al. Visual and auditory neurotoxicity in patients receiving subcutaneous desferrioxamine infusions. New Engl J Med 1986;314:869–73. [10] Porter J, Jawson M, Huehns East C, Hazell J. Desferrioxamine ototoxicity: evaluation of risk factors in thalassaemic patients and guidelines for safe dosage. Br J Haematol 1989;73:403–9. [11] Wonke B, Hoffbrand A, Aldouri M, Wickens D, Flynn D, Stearns M, et al. Reversal of desferrioxamine induced auditory neurotoxicity during treatment with Ca-DTPA. Arch Dis Child 1989;64:77–82.
[12] Aarabi B, Haghshenas M, Rakeii V. Visual failure caused by suprasellar extramedullary hematopoiesis in beta thalassemia: case report. Neurosurgery 1998;42:922–5. [13] Cohen A, Martin M, Mizanin J, Konkle D, Schwartz E. Vision and hearing during deferoxamine therapy. J Pediatr 1990;117: 326–30. [14] Wong V, Li A, Lee A. Neurophysiologic study of beta-thalassemia patients. J Child Neurol 1993;8:330–5. [15] Incorpora G, Di Gregorio F, Romeo M, Pavone P, Trifiletti R, Parano E. Focal neurologic deficits in children with beta-thalassemia major. Neuropediatrics 1999;30:45–8. [16] Georgas D, Paraskevopoulos I, Bezebegis I, Giannitsas N. Weschler Intelligence Scale for Children-III-Manual. Greek ed. Athens: Hellenic Letters; 1997. [17] Cai F, Zhang J. Study of nerve conduction and late responses in normal Chinese infants, children and adults. J Child Neurol 1997;12: 13–18. [18] Gelmi C, Borgna-Pignatti C, Franchin S, Tacchini M, Trimarchi F. Electroretinographic and visual evoked potential abnormalities in patients with beta-thalassemia major. Ophthalmologica 1988;196: 29–34. [19] Bentur Y, Koren G, Tesoro A, Carley H, Olivieri N, Freedman M. Comparison of deferoxamine pharmacokinetics between asymptomatic thalassemic children and those exhibiting severe neurotoxicity. Clin Pharmacol Ther 1990;47:478–82. [20] Amabile G, Stefano E, Bianco I, Aliquo M, Giunti P, Pierelli F. Electrophysiological study in patients with beta-thalassemia major. Acta Neurol 1987;87:181–90. [21] Zafeiriou D, Kousi A, Tsantali C, Kontopoulos E, AugoustidouSavvopoulou P, Tsoubaris P, et al. Neurophysiologic evaluation of long-term desferrioxamine therapy in beta-thalassemia patients. Pediatr Neurol 1998;18:420–4. [22] Papanastasiou D, Papanicolaou D, Magiakou A, Beratis N, Tzebelikos E, Papapetropoulos T. Peripheral neuropathy in patients with beta-thalassemia. J Neurol Neurosurg Psychiatry 1991;54: 997–1000. [23] Davidson C, Weschler I. Erythroblastic (Cooley’s) anemia and neurologic complications. Am J Dis Child 1939;58:362–5. [24] Matoth Y, Shamir Z, Freunollich E. Thalassemia in Jews from Kurdistan. Blood 1955;10:176–8. [25] Logothetis J, Haritos M, Constantoulakis M, Economidou J, Augoustaki O, Lowenson R. Intelligence and behavioral patterns in patients with Cooley’s anemia; a study based on 138 consecutive cases. Pediatrics 1971;48:740–4. [26] Monastero R, Monastero G, Ciaccio C, Padovani A, Camadra R. Cognitive deficits in beta-thalassemia major. Acta Neurol Scand 2000;102:162–8.