Long-term exclusive zinc monotherapy in symptomatic Wilson disease: Experience in 17 patients


  • Potential conflict of interest: Nothing to report.


Exclusive monotherapy with zinc in symptomatic Wilson disease is controversial. Seventeen symptomatic patients with Wilson disease were treated with zinc only. The mean age at diagnosis and start of treatment was 18 years (range 13–26) with approximately half presenting as adolescents. Presentation was exclusively hepatic, exclusively neurologic, and combined in seven, five, and five patients, respectively. The median follow-up was 14 years (range 2–30). At baseline, two of the 12 patients with hepatic disease exhibited decompensated cirrhosis, five exhibited compensated cirrhosis, and five had less severe disease. Both patients with decompensated cirrhosis improved to a compensated state after initiation of therapy. Two of the five patients with initial compensated cirrhosis progressed to decompensated state, and three remain stable. Three of the five patients with moderate or mild liver disease remain stable and two improved. Apart from decreasing bilirubin levels, no significant changes occurred in the liver biochemistry or function during long-term follow-up. Nine of 10 neurologic patients improved markedly and one deteriorated. Two patients with exclusively neurologic presentation developed liver disease during zinc treatment. Two patients with exclusively hepatic presentation developed mild neurologic symptoms. According to 24-hour urinary copper excretions (213 ± 38 versus 91 ± 23 μg: P = 0.01) and serum non–ceruloplasmin-bound copper concentrations (11 ± 2 versus 7 ± 1 μg/dL: P = 0.1) at the end of follow-up, the efficacy of decoppering was less in the exclusively hepatic than in the neurologic group. The prescribed zinc dose and 24-hour urinary zinc excretions tended to be less in the exclusively hepatic group. Conclusion: The outcome of exclusive zinc therapy is generally good in cases of neurologic disease. A less satisfactory outcome in hepatic disease may relate to less efficient decoppering. (HEPATOLOGY 2009.)

Wilson disease is an autosomal recessive disorder of copper transport. Mutations in the ATP7B gene, encoding for a metal-transporting P-type ATPase that is mainly expressed in hepatocytes and functions as a transmembrane copper transporter, lead to imperfect formation of holoceruloplasmin, decreased copper secretion into bile, and thus accumulation of this metal in liver cells and other organs. Symptomatic patients develop hepatic and/or neurologic disease. Despite major advances in understanding the disease pathogenesis at a molecular level, there is still significant controversy on the optimal medical therapy, especially in symptomatic patients. Although initial D-penicillamine treatment is generally recommended under these circumstances, numerous, often severe, side effects can occur during therapy with this chelating agent, requiring discontinuation in up to 30% of cases.1, 2 Especially, neurologic deterioration may occur in a substantial proportion of patients during the initial phase of treatment with D-penicillamine.1–4 As an alternative, treatment with trientine or tetrathiomolybdate has been suggested, but experience with these chelating agents is less extensive. At our center, considerable experience has existed since the 1970s with zinc therapy for Wilson disease. Its mode of action differs from the chelating agents in that it induces enterocyte metallothionein, thus inhibiting copper uptake from the intestinal tract. Because copper also enters the intestinal tract from saliva and gastric secretion, zinc treatment can even generate a negative copper balance.5 Side effects are infrequent, apart from gastric irritation, largely depending on the zinc salt employed. Zinc is often used as the initial, first-line monotherapy in asymptomatic or presymptomatic patients, as well as in symptomatic patients as maintenance treatment after initial decoppering with D-penicillamine.6–12 Although case reports have indicated its efficacy even in patients with decompensated cirrhosis,13 no large series on symptomatic patients treated exclusively with zinc are available. In the current study, we report our experience with zinc monotherapy in symptomatic patients with Wilson disease and present a long-term follow-up.


ANOVA, analysis of variance; CT, computed tomography; MRI, magnetic resonance imaging; PCR, polymerase chain reaction.

Patients and Methods


Of the 57 patients with definite symptomatic Wilson disease in whom therapy was started in our hospital after 1975, zinc was the initial therapy of choice in 23 cases. Initial results of zinc therapy in a subgroup of these patients were reported in 1987.8 In 17 patients, baseline and long-term follow-up data were available, and these patients are reported here. Patients visited the outpatient department on average twice a year. All 17 patients were treated after diagnosis with exclusive zinc therapy and their data are included until the latest follow-up (2008 in 14 patients) or until zinc treatment was replaced or supplemented with other medication for their Wilson disease (one patient), lost to follow-up (one patient), or underwent liver transplant (one patient). The zinc dose was modified based on clinical symptoms and signs, serum non–ceruloplasmin-bound copper concentrations (aim: <10 μg/dL), and urinary copper excretions (aim: < 100 μg/24 hours) during follow-up.

Liver disease severity at baseline was classified as: (1) mild (disturbed liver biochemistry only); (2) moderate (disturbed liver biochemistry combined with discrete abnormalities on ultrasound, computed tomography [CT] scan or magnetic resonance imaging [MRI], and/or, if performed, slight or moderate fibrosis on liver biopsy but without any laboratory, radiologic, or histologic abnormalities suggesting cirrhosis); (3) compensated endstage liver disease (cirrhosis evident by clinical, laboratory, radiologic, or histologic findings, classified as stage A according to the Child-Pugh score); or (4) decompensated end-stage liver disease (cirrhosis evident by clinical, laboratory, radiologic, or histologic findings, classified as stage B or C according to the Child-Pugh score). At end of follow-up, liver disease was categorized as improved, stable, or deteriorated depending on changes in the liver disease severity score from baseline. Neurologic symptoms and signs were described as parkinsonian syndrome, ataxic syndrome, or dystonic syndrome. Tremor and other symptoms were recorded separately. Speech was scored on the speech function scale (0 indicates normal, 7 indicates anarthria).6 At the end of follow-up, all patients were also scored by an experienced neurologist (F.H.H.L.) on the modified Rankin scale (grade 0 is no handicap; grade 1 is minor symptoms that do not interfere with lifestyle; grade 2 is minor handicap; grade 3 is moderate handicap; grade 4 is moderately severe handicap; grade 5 is severe handicap, totally dependent, requiring constant attention day and night).14 At latest follow-up, clinical symptoms and signs as well as the ancillary investigations were classified as improved, stable, or worse. All available MRI scans of the brain were reviewed by one investigator (F.H.H.L.). Abnormal gray matter nuclei were localized by tracing the nuclei on the T1-weighted inversion-recovery MR images and noting the abnormal changes in the increased signal intensity (SI) on the T2-weighted SE images of the MRI in the basal ganglia, thalamus, and dentate nucleus. Abnormal white matter signal intensity as well as abnormal white matter signal intensity specific in the corticospinal tract, dentatothalamic tract, or pontocerebellar tract were noted. Atrophy of the brain was scored separately.15

For each patient we include data at baseline, at the end of follow-up, and during the entire follow-up for 24-hour urinary copper and zinc excretions, serum non–ceruloplasmin-bound copper concentrations, and dose elemental zinc prescribed. For laboratory values, numbers of determinations during each year varied considerably (up to six determinations per year). To avoid bias, we first calculated mean values for each year, and subsequently from these mean values, average values during the entire follow-up for each patient (Table 3). All laboratory data from 1992 onward were retrieved from the electronic database introduced that year in our hospital. Laboratory data before 1992 were retrieved from laboratory reports contained in the patient charts. Clinical data were obtained from patient charts and correspondence. Baseline non–ceruloplasmin-bound copper concentrations could not be retrieved in two patients, and baseline serum ceruloplasmin levels and 24-hour urinary copper excretion could not be retrieved in one patient each. Despite extensive efforts, follow-up data on 24-hour urinary copper/zinc excretions and serum non–ceruloplasmin-bound copper concentrations before 1992 could not be retrieved in five patients and one patient, respectively. Data on baseline urinary copper excretion and urinary copper/zinc excretion during follow-up from 2000 on are always based on 24-hour collections.1, 2 Before 2000, follow-up data on urinary copper/zinc excretion were often based on urine aliquots retrieved during visits to the outpatient clinic. In these cases we estimated 24-hour urinary copper/zinc excretions, assuming 24-hour urinary volumes of 1500 mL. In one neurologic patient, 24-hour urinary copper/zinc excretion could not be obtained during the last 5 years of follow-up because of urinary incontinence.

Table 3. Efficacy of Decoppering and Clinical Outcome During Zinc Monotherapy in 17 Patients with Wilson Disease.
Patient No.FU (years)24-hour Urinary Copper (μg)Serum Copper (μg/dL)*Elemental Zinc Dose (mg)24-hour Urinary Zinc (μg)Baseline/End ALT (U/L)Outcome
Normal <40Normal <15  Normal <35
  • H, C, N indicate hepatic, combined, or neurologic presentation. FU, follow-up; NA, not available. Laboratory values and zinc dose are given as mean ± SEM or in case of nonparametric distribution, as median and range during entire follow-up period. In patients with hepatic or neurologic deterioration during follow-up, laboratory values and zinc dose preceding the deterioration are also given in parentheses.

  • *

    Serum non–ceruloplasmin-bound copper concentration.

  • Elemental zinc dose prescribed.

1 H11299 ± 576 ± 1136: 68–1366821 ± 791120/26Normalized aminotransferases.
2 H5231 ± 810 ± 2168: 136–2501336 ± 8165/226Increasing aminotransferases, persistent coarse pattern and steatosis on ultrasound
3 H11162 ± 209 ± 2136: 100–1361135 ± 338125/84Persistent abnormal aminotransferases (suboptimal therapy compliance)
4 H25358 ± 32 (546)16 ± 2 (24)136: 45–363 (45)2325 ± 390 (NA)18/82 (23)Stable compensated cirrhosis Mild neurological symptoms after 10 years (after zinc dose reduction)
5 H20254 ± 47 (315)10: 1–46 (40)184: 136–276 (207)1151: 690–5720 (690)24/16 (8)Initial decompensated cirrhosis improved to compensated cirrhosis. Mild neurologic symptoms after 12 years (period of uncompliance).
6 H16155 ± 1910 ± 1207: 69–2765249 ± 71524/32Stable compensated cirrhosis.
7 H2360: 360–74118: 9–382762443: 2342–574563/123Initial decompensated cirrhosis improved to compensated cirrhosis.
8 C15270: 180–750 (750)7: 2–45 (45)207: 69–276 (207)3728: 1730–4500 (2252)32/47 (44)Strong improvement of neurologic symptoms. Progression to decompensated cirrhosis, liver transplant. Poor compliance.
9 C24102 ± 9 (76)1: 1–5 (3)138: 92–107 (92)6773: 1155–14437 (3320)32/48 (48)Strong improvement neurologic symproms. Gradual progression from compensated to decompensated cirrhosis:evaluation transplant
10C441: 39–1877 ± 2276: 184–5524236 ± 92945/59Strong improvement neurologic symptoms
       Persistent slight elevation aminotransferases and coarse pattern of liver at ultrasound.
11N11122 ± 118 ± 1207: 207–2762274 ± 23811/17Strong improvement neurologic symptoms
12C1796 ± 126 ± 1207: 138–2076355 ± 91459/57Strong improvement neurologic symptoms Initial coarse pattern on ultrasound normalized at follow up. Mild fibrosis on baseline liver biopsy, normal liver biopsy 3 years later. Persistently disturbed aminotransferases.
13C12254 ± 457 (3–31)207: 207–3682828: 1230–964521/43Strong improvement of neurologic symptoms Stable compensated cirrhosis.
14N30148 ± 15 (250)4: 1–24 (8)207: 138–276 (276)3684: 1170–30083 (3435)31/19 (51)Severe deterioration neurologic symptoms. Development compensated cirrhosis.
15N7116 ± 125 ± 1153: 138–2073553 ± 27515/35Strong improvement neurologic symptoms
16N893 ± 18 (66)12 ± 1 (18)207: 200–276 (207)2724 (1272–11325) (1631)20/111 (122)Strong improvement neurologic symptoms. Development of increased aminotransferases and coarse pattern on ultrasound (initial normal)
17N29182: 32–4138 ± 1276: 200–4606795: 2280–80840NA/20Strong improvement neurologic symptoms


Ceruloplasmin was measured by nephelometry (Behring Nefelometer II; Dade Behring, Eschborn, Germany): lower limit of detection 2 mg/dL, coefficient of variation 2.2% at a concentration of 17.5 mg/dL and 3.5%-4.2% at concentration 41.3 mg/dL. Concentrations of copper and zinc in serum and urine were measured by atomic absorption spectrophotometry (SpectrAA220 Varian, Palo Alto, CA; wavelength for copper 324.8 nm and for zinc 213.9 nm). Twenty-four-hour urinary excretions were collected in copper- and zinc-free containers and concentrations determined after acidification. Serum non–ceruloplasmin-bound copper concentrations were calculated as the difference between serum copper concentrations (in μg/dL) and 3.15 times the ceruloplasmin concentrations (in mg/dL) at the same timepoint. Screening for the H1069Q mutation was performed in all patients. Polymerase chain reaction (PCR) was performed using the forward primer AGT TCT GCC TCA GGA GTG TGA C and the reverse primer CAG CTA CCA GAG AAG GAC ATG to amplify exon 14 of the ATP7B gene from genomic DNA samples. The resulting product was subjected to sequencing using the Big Dye Terminator Cycle Sequencing Ready Reaction Mix (Applied Biosystems, Foster City, CA). Samples were analyzed on an ABI 3700 automatic sequencer. In each set of PCR reactions control DNA from a known H1069Q homozygote and a heterozygote were amplified and sequenced under identical conditions.

Statistical Analysis.

Values are given as means ± standard error of the mean (SEM) or in case of non-Gaussian distribution, as medians and range. Differences between various timepoints or between groups were analyzed for significance by analysis of variance (ANOVA) with Fisher's LSD test as a post-hoc test, unpaired or paired t tests, or Mann-Whitney U tests, as appropriate. A two-sided P-value < 0.05 was considered statistically significant.


Baseline Patient Characteristics.

Baseline clinical and laboratory features for the exclusively hepatic and neurologic groups are given in Table 1 and for the individual patients in Table 2. The mean age at diagnosis and start of treatment was 18 years (range 13-26) with approximately half presenting as adolescents. Many patients exhibited very low serum ceruloplasmin levels, suggesting severe mutations. Presentation was exclusively neurologic in five patients, exclusively hepatic in seven patients, and combined neurologic and hepatic in five patients. The median time from diagnosis to end of follow-up was 14 years (range 2-30 years), without difference between various presentation modes.

Table 1. Frequency of Clinical and Laboratory Criteria Used for Diagnosis in 17 Patients with Wilson Disease
CharacteristicNeurologicExclusively HepaticAll
  • *

    One basal value missing.

  • Two basal values missing.

  • Performed in seven patients.

Patient no.10717
Median age (years) at diagnosis (range)21 (13–26)19 (13–23) 
Kayser-Fleischer rings10/10 (100%)3/7 (43%)13/17 (76%)
Ceruloplasmin <20 mg/dL9/9 (100%)*7/7 (100%)16/16 (100%)
Serum non–ceruloplasmin-bound copper conc >15 μg/dL8/8 (100%)7/7 (100%)15/15 (100%)
Urinary 24-hour copper >100 μg9/9 (100%)*7/7 (100%)16/16 (100%)
Liver biopsy (Cu >250 μg/g dry weight)5/6 (83%)1/1 (100%)6/7 (86%)
Table 2. Baseline Clinical and Laboratory Data in 17 Patients with Wilson Disease
Patient No.AgePresentationKayser-Fleischer RingCeruloplasmin (mg/dL)Serum Non–Ceruloplasmin-Bound Copper (μg/dL)24-hour Urinary Copper (μg)Liver BiopsyMutation
  • *

    Copper in μg/g dry weight

  • Baseline values not available in one patient.

  • Baseline values not available in two patients.

  • §

    Biopsy performed after improvement to compensated cirrhosis

  • CAH, chronic active hepatitis; CU, copper staining.

113HepaticNo1617825 H1069Q/
223HepaticNo518216Steatosis, no fibrosis: CU– 
419HepaticYes517497Cirrhosis, CAH: CU+H1069Q/
515HepaticYes1337540§Cirrhosis, CAH: CU+: 800*H1069Q/H1069Q
622HepaticYes1130270Cirrhosis, CAHH1069Q/H1069Q
719HepaticNo13595,100§Cirrhosis, CAH: CU+ 
826CombinedYes1520 Severe fibrosis: 1100*H1069Q/H1069Q
925CombinedYes  220Cirrhosis, CAH: CU+: 945* 
1121NeurologicYes627235Normal: 554*H1069Q/
1221CombinedYes6481,335Mild fibrosis, inflam: CU+: 397*H1069Q/H1069Q
1316CombinedYes10271,455Severe fibrosis, CAH: CU–H1069Q/
1413NeurologicYes454430Normal: 1100* 
1621NeurologicYes528299 H1069Q/
1714NeurologicYes3 390No fibrosis: 135* 

All patients with neurologic symptoms displayed Kayser-Fleischer rings at baseline, which was the case in only half of the patients with exclusively hepatic presentation. Serum ceruloplasmin levels were decreased, whereas 24-hour urinary copper excretions and serum non–ceruloplasmin-bound copper levels were elevated in all patients with available baseline values. The results of histological examination of baseline liver biopsies (obtained in 12 patients) are given in Table 2. Copper content in liver biopsy (determined in seven patients) was >250 (range 270–1,100) μg/g dry liver weight in six cases (86%) and in one case with a value of 135 μg/g dry liver weight, still suggestive of Wilson disease according to recent American Association for the Study of Liver Diseases (AASLD) guidelines.2 Repeated biopsy 4 years later in this patient revealed a copper concentration of 335 μg/g dry liver weight. Four patients (23%) were homozygous and five patients (29%) heterozygous for the H1069Q mutation in the ATP7B gene (Table 2).

Details of Zinc Monotherapy.

In all patients, zinc was initially prescribed as sulfate. Generally, the initial dose after diagnosis was 135 mg elemental zinc/day. In four patients a higher initial dose (207 mg elemental zinc daily in three cases, and 276 mg elemental zinc in one case) was initially prescribed because significant clinical problems were perceived. In two children the initial dosage of elemental zinc was 69 mg. The mean daily elemental zinc dose was 158 ± 14 mg (median and range 138: 69–276) at baseline and 199 ± 16 mg (200: 92–350) at the end of follow-up (P = 0.03 compared to baseline), without differences between the 10 patients with neurologic or the seven with exclusively hepatic presentation. At end of follow-up, median 24-hour urinary zinc excretion was 3710 μg (range 1,200–34,534), without significant differences between the 10 patients with neurologic or the seven with exclusively hepatic presentation. During the entire follow-up, average elemental zinc dose (208 ± 13 versus 177 ± 19 mg/day: P = 0.08) and average urinary zinc excretion (4,295 ± 543 versus 2,922 ± 845 μg: P = 0.11) tended to be higher in the neurologic group. Nevertheless, we could find no clear relationship between dose or 24-hour urinary excretion of zinc with a favorable or unfavorable course of the disease. Side effects of zinc therapy were infrequent, except gastrointestinal complaints in six patients, generally improving or resolving after substitution of zinc sulfate by zinc acetate (five patients) or zinc gluconate (one patient). In one patient (patient 14), severe granulocytopenia and macrocytic anemia occurred after 30 years of zinc therapy, potentially related to the zinc treatment, or to concurrent antiepileptic drugs (phenytoin and carbamazepine). These problems turned out to be reversible after temporary interruption of all medication.

Effects of Zinc Treatment on Hepatic Disease.

Duration of follow-up in the 12 patients with initial hepatic presentation was 13 years (median; range 2–25 years). Laboratory data and clinical outcome during follow-up are given for each patient in Table 3. Two patients exhibited severely decompensated cirrhosis at baseline. The first (patient 7), an 18-year-old female, presented with ascites, hemolysis (Hb 5.3 mMol/L, LDH 3535 IU/L, reticulocytes 181‰ [normal <30‰], total bilirubin 4.4 mg/dL [conjugated 3.6 mg/dL], albumin 23.1 g/L, prothrombin time 17.7 seconds [control 12.5 seconds]). Ultrasound revealed cirrhosis and ascites. The second patient, a 15-year-old male (patient 5), presented with hemolysis (Hb 4.7 mMol/L), total bilirubin 8.8 mg/dL (conjugated 3.6 mg/dL), albumin 22.6 g/L, and antithrombin level of 9%. Ultrasound revealed cirrhosis and collateral circulation, consistent with portal hypertension. Both improved to compensated cirrhosis within a year after initiation of therapy, with persistently normal parameters of liver synthetic function after 2 and 19 years of follow-up. Five patients presented with compensated cirrhosis. Two of these progressed to a decompensated state after 15 and 24 years of zinc therapy. The first patient (patient 8) exhibited unexplained neurologic symptoms since 1976, at an age of 21 years. The diagnosis of Wilson disease was established in 1981, and zinc treatment prescribed. Liver biopsy at that time revealed severe fibrosis, whereas CT scan and thrombocytopenia (105 × 109/L) indicated cirrhosis. A follow-up CT scan in 1990 revealed progressive cirrhosis, with a nodular aspect of the liver and collateral circulation, and a repeat liver biopsy in 1992 also revealed cirrhosis. In the following years, compliance with the prescribed medication was admitted to be poor. She was transplanted in 1995 because of deterioration of liver synthetic parameters and occurrence of ascites. The second (patient 9) was referred for liver transplant because of gradual deterioration of liver synthetic parameters (including albumin decreasing to 17 g/L) during a period of 5 years and occurrence of diuretic-responsive ascites, despite apparently sufficient decoppering for many years. The other three patients with compensated cirrhosis remain stable at end of follow-up after 25, 16, and 12 years of zinc therapy. Of the three patients with initially moderate liver disease, two remain stable and one improved. Of the two patients with initially mild liver disease, one remains stable and in the other liver biochemistry normalized. In the 12 patients with hepatic disease at presentation, serum bilirubin decreased during follow-up due to the improvement in the two patients with initial hemolysis, but there was no significant change in liver biochemistry, albumin, prothrombin time, or thrombocytes (Table 4). During follow-up two additional patients with exclusively neurologic presentation developed hepatic disease: the first (patient 14), described in detail below, with severe neurologic deterioration during zinc therapy, developed compensated cirrhosis under zinc treatment. He had normal liver biopsy at baseline, but pronounced fibrosis at repeated biopsy after 7 years (copper content 1,100 and 270 μg/g dry liver weight in baseline and follow-up biopsies). He gradually deteriorated neurologically in the next 4 years (see below) in combination with portal hypertension and collateral circulation on CT scan and ultrasound. Liver synthetic function remained adequate during the further follow-up (total 30 years). Because of urinary incontinence, no data on 24-hour urinary zinc or copper excretion during the last 5 years of follow-up are available. The second patient (patient 16) exhibited normal liver biochemistry and ultrasound at baseline, but developed pronounced increases in serum aminotransferases (ALT 140 U/L) and a coarse pattern of the liver at ultrasound after 6 years of zinc therapy.

Table 4. Liver Biochemistry and Parameters of Synthetic Function During Zinc Therapy in 12 Patients with Wilson Disease Presenting with Hepatic Disease and Five Patients with Wilson Disease Presenting with Exclusively Neurologic Disease
 BaselineEnd of Follow-UpP ValueNormal Values
  • Data are given as means ± SEM, and in case of nonparametric distribution in parentheses also as median and range.

  • *

    Baseline values missing in one patient.

Hepatic (FU: years) 13 (2–25)  
Bilirubin (mg/dL)1.9 ± 0.9 (0.6: 0.3–8.8)0.9 ± 0.1 (0.8: 0.5–1.7)0.23< 1.2
Alk Phosph (U/L)125 ± 29144 ± 150.55< 120
γ-glutamyltransferase (U/L)69 ± 1684 ± 360.71< 55
AST (U/L)47 ± 945 ± 90.88< 35
ALT (U/L)54 ± 1167 ± 170.77< 45
Albumin (g/L)38 ± 3 (42: 22–48)38 ± 2 (38: 22–48)0.8535–50
Prothrombin time (seconds)14.7 ± 0.6 (14.9: 12.5–17.7)14.4 ± 0.2 (14.2: 13.4–16.1)0.7312.0–15.0
Thrombocytes (109/L)146 ± 23 (108: 80–327)157 ± 27 (134: 36–347)0.76150–450
Neurologic*(FU: years) 11 (7–30)  
Bilirubin (mg/dL)0.8 ± 0.20.9 ± 0.20.64 
Alk Phosph (U/L)107 ± 22103 ± 140.88 
γ-glutamyltransferase (U/L)28 ± 434 ± 60.46 
AST(U/L)20 ± 318 ± 20.54 
ALT (U/L)14 ± 240 ± 18 (20: 17–111)0.23 
Albumin (g/L)42 ± 142 ± 20.73 
Prothrombin time (seconds)13.2 ± 0.514.6 ± 0.30.06 
Thrombocytes (109/L)159 ± 17205 ± 240.20

Effects of Zinc Treatment on Neurologic Disease.

The duration of follow-up in the 10 patients with neurologic disease (five with concurrent hepatic disease) was 13 years (median; range 4–30 years). Laboratory data and clinical outcome during follow-up are given for each patient in Table 3. One patient (patient 14) experienced severe neurologic deterioration (described below), and nine patients improved neurologically during zinc monotherapy. The results in these patients were as follows: Dysarthria (always present at baseline) improved considerably in all, and was only minor at end of follow-up in six patients and completely disappeared in three patients. Tremor (present at baseline in six patients) was very mild at the end of follow-up in three patients and absent in the other three patients. Hypokinetic rigid syndrome and dystonia (present in four and three patients at baseline) disappeared in all. Ataxia was present in one patient at baseline but disappeared completely during zinc therapy. Improvement of various neurologic symptoms occurred within 1-24 months (median 12 months). The Rankin score at the end of follow-up was 1 in all cases. All these patients have a job and live independently. One of these patients needs supervision because of psychiatric symptoms. In contrast, as mentioned above, one patient (patient 14) experienced severe deterioration. He presented with exclusively neurologic disease (mild dysarthria and dystonia) in 1977. Zinc was prescribed from the beginning, but patient compliance with this treatment was initially poor. From 1981 on he deteriorated progressively in 4 years, despite good compliance with medication at that time, disappearance of the Kayser-Fleischer rings, and adequate serum non–ceruloplasmin-bound copper levels (8 μg/L) but clearly elevated 24-hour urinary copper excretion (250 μg). He also developed progressive hepatic disease in that period, with compensated cirrhosis (see above). At end of follow-up, serum non–ceruloplasmin-bound copper concentration was 2 μg/dL, 24-hour urinary copper excretion not available because of incontinence. From a neurologic point of view he remains severely handicapped, with a current Rankin score of 5 and dysarthria score of 6-7 and need for antiepileptic medication because of seizures. In addition to the 10 patients presenting initially with neurologic disease, two of the seven patients with baseline exclusively hepatic presentation developed new neurologic symptoms after 10 and 12 years of zinc therapy. Both had a Rankin score of 1 at the end of follow-up. The first patient (patient 4) exhibited tremor and changed handwriting (micrographia) within a few months after an inadvertent decrease of the zinc dose prescribed (daily dose decreased by physician in another hospital from 138 to 46 mg elemental zinc). Laboratory values indicated insufficient decoppering during this episode (Table 3). Symptoms improved again after increasing the zinc dose. The second patient (patient 5) exhibited new dysarthria, rigidity, tremor, and dystonia in combination with abnormalities of the MRI brain (normal at baseline) after 12 years of zinc monotherapy. Twenty-four-hour urinary copper and zinc excretion (315 respectively 690 μg) and serum non–ceruloplasmin-bound copper concentration (40 μg/dL) indicated insufficient decoppering and bad compliance with the prescribed treatment of 207 mg elemental zinc, as admitted by the patient. In fact, laboratory parameters and patient history indicated a prolonged period of several years of therapy noncompliance before the neurologic symptoms developed. Neurological symptoms and laboratory parameters improved thereafter, supposedly related to improved compliance.

MRI of the brain was performed in 13 patients (five neurologic, five combined, and three hepatic presentation) within 1 year after diagnosis. At baseline, increased signal intensities on T2-weighted images were found in the basal ganglia (including the putamen, globus pallidus, nucleus caudatus, nucleus lentiformis, and claustrum) in eight patients, in the thalamus in four, and in the mesencephalon in one patient. Abnormalities in signal intensity in the corticospinal tract were observed in five patients and in the pontocerebellar tract in one. Other abnormal white matter signal intensity was found in three patients. Atrophy of the cerebellar lobes was demonstrated in one, and frontal lobe atrophy in two patients. Follow-up MRI was done in 10 patients with a median interval of 7 (range 1-13) years after the baseline investigation. Follow-up MRI showed improvement in four patients (patients 4, 10, 11, 12). In all four, abnormal signal intensities in the basal ganglia, i.e., in the corpus striatum, putamen, and nucleus caudatus disappeared. In three cases, abnormal signal in thalamus, corticospinal, and pontocerebellar tracts also disappeared. Furthermore, subcortical lesions disappeared in one patient. MRI showed no change of abnormalities in four patients (patients 6, 9, 14, 16). Worsening of MRI brain occurred in two patients (patients 5 and 17). One of these (patient 5) presented with decompensated cirrhosis. Although he improved toward compensated liver disease during zinc therapy, he also developed mild neurologic symptoms after 12 years treatment, as mentioned above. Baseline MRI was normal, but follow-up MRI after 12 years of therapy showed extensive increased signal intensity on T2-weighted images in putamen, globus pallidus, thalamus, and corticospinal tract, coinciding with insufficient decoppering. The second patient (patient 17), presenting with neurologic symptoms, showed only slight abnormalities at the right putamen at baseline MRI. Although he improved clinically during zinc therapy, follow-up MRI after 5 years revealed slightly progressive lesions and after 14 years more extensive abnormal signal intensities in putamen, globus pallidus, internal capsule, and mesencephalon, despite adequate decoppering at that time.

Efficacy of Decoppering.

After 1 year of therapy and at the end of follow-up, serum non–ceruloplasmin-bound copper concentrations were decreased significantly compared to baseline values, both in the neurologic and the exclusively hepatic groups (Table 5, Fig. 1). Whereas baseline non–ceruloplasmin-bound copper concentrations did not differ between the groups, values at the end of follow-up tended to be lower in the neurologic group (P = 0.1). In fact, serum non–ceruloplasmin-bound copper concentrations at the end of follow-up remained above 10 μg/dL in two patients (20%) with neurologic presentation and five (71%) with exclusively hepatic presentation. Also, average serum non–ceruloplasmin-bound copper concentrations during the entire follow-up were higher in the exclusively hepatic than in the neurologic group (Table 5, Fig. 1). In the neurologic patients, 24-hour urinary copper excretion decreased significantly after 1 year therapy and at the end of follow-up. In the group with exclusively hepatic presentation, 24-hour urinary copper excretion tended to decrease after 1 year and at the end of follow-up, but significance was not reached. At baseline, 24-hour urinary copper excretion was not different between the groups with neurologic and exclusively hepatic presentations, but average values during the entire follow-up and values at the end of follow-up were significantly higher in the exclusively hepatic group (Table 5, Fig. 2). Differences were also significant when the exclusively hepatic group was compared with the group of five patients with exclusively neurologic presentation (data not shown). At end of follow-up, 24-hour urinary copper excretion was below 100 μg in all four patients with exclusively neurologic presentation (not determined in the fifth because of incontinence), in two of five patients with combined presentation, and in only one of seven patients with exclusively hepatic presentation.

Table 5. Efficacy of Decoppering During Zinc Therapy in 10 Patients with Neurologic Presentation and in Seven Patients with Exclusive Hepatic Presentation of Wilson Disease
 BaselineOne YearEnd Follow-UpAverage Values During Follow-Up
  • Data are given as means ± SEM, and in case of nonparametric distribution also in parentheses as median and range.

  • Missing 2 values at baseline and 1-year follow-up.

  • Missing 1 value at baseline and 1-year follow-up, and (because of urinary incontinence) 1 value at end of follow-up.

  • *

    Significantly different (P = 0.01) from baseline;

  • **

    significantly different (P = 0.01) from neurologic group.

Serum non–ceruloplasmin-bound copper concentration (μg/dL)    
Neurologic28 ± 35 ± 2*7 ± 1*6 ± 1
Hepatic28 ± 58 ± 3*11 ± 2*11 ± 1**
Urinary copper excretion (μg/24-hour)    
Neurologic590 ± 159 (390: 220–1445)157 ± 29* (162: 50–270)91 ± 23* (78: 14–270)142 ± 23 (119: 41–270)
Hepatic1084 ± 674 (497: 144–5100)256 ± 46 (258: 151–360)213 ± 38** (225: 38–360)249 ± 27** (254: 155–358)
Figure 1.

Serum non–ceruloplasmin-bound copper concentrations in individual patients at baseline and during zinc treatment in (A) seven Wilson patients with exclusively hepatic presentation and (B) 10 patients with neurologic presentation (closed symbols, continuous lines, five cases with coexistent hepatic disease: open symbols, interrupted lines, exclusively neurologic presentation). Average annual values during the entire follow-up are given for each patient. Mean annual values for the entire subgroup are indicated by bars.

Figure 2.

Twenty-four-hour urinary copper excretion in individual patients at baseline and during zinc treatment in (A) seven Wilson patients with exclusively hepatic presentation and (B) 10 patients with neurologic presentation (closed symbols, continuous lines, five cases with coexistent hepatic disease: open symbols, interrupted lines, exclusively neurologic presentation). Average annual values during the entire follow-up are given for each patient. Mean annual values for the entire subgroup are indicated by bars.


The main findings of the current study in symptomatic patients with Wilson disease on long-term zinc treatment without initial chelating therapy are as follows. First, results for symptomatic neurologic disease are highly satisfactory. Second, preexistent hepatic disease often remains stable and may even improve to a compensated state in patients with decompensated cirrhosis. Third, preexisting liver disease may also deteriorate, with potentially severe consequences. Fourth, this approach does not always prevent new liver disease in patients with exclusively neurologic presentation. Last, exclusive zinc therapy is associated with few side effects apart from gastrointestinal complaints. Zinc therapy improved neurologic symptoms and signs remarkably in 9 of 10 patients with neurologic presentation. Neurologic disease worsened significantly in one patient, and emerged in two patients with exclusively hepatic presentation during periods of poor compliance and inadvertent dose reduction.

Clinical outcome was less satisfactory for hepatic disease. Although two patients with decompensated cirrhosis experienced a remarkable recovery to a compensated state, two other patients with initially compensated cirrhosis deteriorated during treatment, with need for liver transplant. Also, the fact that two patients with initial exclusively neurologic disease developed hepatic disease during zinc monotherapy underlines the contention that close hepatologic observation is essential during zinc treatment. We also observed progression to a decompensated state in another patient with initial compensated cirrhosis on zinc monotherapy (not included in the present study because she initially received penicillamine for decoppering). This patient improved markedly, with compensated liver disease reestablished after adding trientine to the zinc therapy.

Zinc therapy induced less effective decoppering in the hepatic group than in the neurologic group, when judged by 24-hour urinary copper excretion and serum non–ceruloplasmin-bound copper concentrations (Table 5, Figs. 1, 2). Zinc doses prescribed and 24-hour urinary zinc excretion tended to be lower in the exclusively hepatic than in the neurologic group. The value of this finding is uncertain. One might speculate that in the case of neurologic symptoms, patient compliance could be better, and physicians could be inclined to prescribe higher doses because they were directly confronted with the bothersome neurologic symptoms. On the other hand, the value of 24-hour urinary zinc excretion is limited, because patients can increase zinc ingestion shortly before visiting the clinic in order to conceal poor compliance. In the current study, periods of noncompliance and/or deterioration were often not associated with particularly low 24-hour urinary zinc excretion, and we could find no clear relationship between dose or 24-hour urinary zinc excretion with a favorable or unfavorable course of the disease. Also, in some patients (patients 8, 9, 16) neurologic symptoms improved, whereas the coexistent liver disease worsened. Nevertheless, we cannot exclude that higher zinc doses and/or better compliance with prescribed treatment could have led to a more favorable outcome in the hepatic group.

In the Netherlands, zinc has been used for many years since its introduction by Schouwink in 1961.8 There are only limited data on exclusive zinc therapy in symptomatic Wilson disease.7, 8, 16 We recently performed a systematic review on the clinical efficacy of chelator agents and zinc in the initial treatment of Wilson disease.17 The results of our systematic review were remarkably similar to the current data, with general improvement for neurologic symptoms, but much less favorable results for patients with hepatic disease. A major advantage of our study is the relatively long follow-up (median 14 years, range 2–30), much longer than the follow-up in the previous reports. Also, we were able to present various relevant clinical details in this single-center study. The major disadvantage of our study (as well of all previous reports) is its retrospective character. Another limitation of the current work is that before 2000, follow-up data on urinary copper and zinc excretion were often based on urine aliquots. Although this policy is currently considered suboptimal,1, 2 we also include these data to avoid potential bias: Before 2000, 24-hour urinary collections may in theory have been performed especially when there was a clinical reason for it (e.g., neurologic or hepatic deterioration, elevated serum non–ceruloplasmin-bound copper concentrations, or suspicion of noncompliance). Nevertheless, the results on urinary copper excretion were essentially the same when only data from 24-hour urinary collections were taken into account (results not shown). A placebo-controlled randomized trial would be preferable to demonstrate equivalence, inferiority, or superiority of zinc treatment compared to chelators, but this is difficult to achieve in this rare disease. In the clinical situation where it is unfeasible to perform randomized controlled trials, careful retrospective data may yield evidence relevant for clinical decision-making. Based on the available literature, it is our opinion that zinc monotherapy is the therapy of choice in asymptomatic and presymptomatic patients and in patients with exclusively neurologic presentation, considering the significant potential for neurologic deterioration with chelating agents. Zinc monotherapy may also be considered in patients with exclusively hepatic or combined neurologic and hepatic presentation, provided that the liver disease is not advanced. Although zinc monotherapy may yield remarkable results in patients with advanced liver disease, we advise prior chelating treatment for initial decoppering given the results presented here and in the previous literature.17

In conclusion, in this study with relatively long follow-up in patients with Wilson disease, exclusive zinc monotherapy generally yielded satisfactory results for neurologic disease. In the case of hepatic disease the results were less satisfactory, possibly related to less effective decoppering.


We thank Dr. Tjaard U. Hoogenraad, the initiator and promotor of zinc therapy for Wilson disease at the University Hospital Utrecht, The Netherlands, and Dr. Michele Vacca for expert assistance with the figures.