What Is Known
- The diagnosis of Wilson disease is difficult because of the heterogenous manifestations, especially in children.
- The most specific diagnostic test for Wilson disease is ATP7B mutation analysis.
- Alkaline phosphatase, a zinc (Zn)-dependent enzyme is usually low in severe Wilson disease patients; this may be explained by low serum Zn.
What Is New
- Serum Zn is valuable in diagnosing Wilson disease.
- Serum Zn correlates with the severity of Wilson disease.
- A proposed diagnostic score that includes serum Zn in the equation, could help diagnosis and early initiation of treatment while genetic results are awaited or not available.
Wilson disease (WD) is an inherited disorder of copper metabolism with diverse clinical manifestations. There are many atypical presentations making the diagnosis difficult (1–3). The genetic defect is attributed to a mutation in the ATP7B gene, which mainly impairs the excretion of copper from hepatocytes into bile (4). Early diagnosis is important so treatment can be initiated in a timely manner, before the patient advances to irreversible liver failure or neurological damage.
Consensus diagnostic criteria for WD were formalized at the 8th International Meeting on Wilson disease, Leipzig in 2001 and published by Ferenci et al (5). The diagnostic score was retrospectively validated in children and showed to have a high specificity (96.6%) and sensitivity (98.1%) (6). This diagnostic score is widely used (7); however, it includes ATP7B analysis that may be unavailable in many resource-limited settings (8) or may take several weeks before the result is available. Unfortunately, the genetic mutation cannot unequivocally confirm the diagnosis of WD because there are more than 600 mutations located on the ATP7B gene and not all are disease-causing mutations (9). Consequently, we sought for alternative parameters, which are more readily available in resource-limited settings, to replace the ATP7B mutation analysis.
Zinc (Zn) is used to treat WD because Zn can inhibit copper absorption in the intestine by up regulation of metallothionines (MTs) in both enterocytes and hepatocytes leading to a decrease of copper absorption and copper toxicity in WD patients (10). The mechanism of Zn as a decopper agent has been known for a long time; however, no studies have focused on the role of serum Zn in untreated WD patients.
In this study, we hypothesized that there was a correlation between low serum Zn and the severity of WD. We proposed a disease diagnostic score that included serum Zn as a novel parameter (Proposed WD diagnostic score) and retrospectively validated its diagnostic value.
Medical records of WD and non-WD children seen at King's College Hospital (KCH), London, were retrospectively reviewed from 2005 to 2015. The demographic data and laboratory investigations were collected and only the data from the initial presentation at KCH were analyzed. All WD children (n = 43) had an ATP7B mutation confirmed. The WD patients were classified into 2 subgroups: ALF and non-ALF based on the definition by the Pediatric Acute Liver Failure study group (11). The Revised WD index as published by Dhawan et al (12) was calculated in WD children who had initial serum Zn data (n = 26).
To find the new parameters and revise the cutoff level of the original parameters from the WD diagnostic score published by Ferenci et al (5), data from WD (n = 43) and non-WD (n = 375) were compared and critically evaluated. Eligible subjects in the non-WD group were children with a new presentation of autoimmune hepatitis (AIH) (n = 140), nonalcoholic fatty liver disease (NAFLD) (n = 174) and ALF from other causes (n = 61; toxic hepatitis [n = 7], infectious etiology [n = 7], metabolic liver disease but not WD [n = 13], ischemic liver [n = 1], infiltrative disease [n = 2], AIH [n = 1], and indeterminate etiology [n = 30]).
To validate our proposed diagnostic score (Proposed WD diagnostic score), we studied AIH (n = 63) and NAFLD patients (n = 59) who had the complete workup for all parameters required by the WD diagnostic score published by Ferenci et al (5) as the control groups to reduce the number of false-negative results and selection bias.
Laboratory Data Collection
Copper measurement, Rhodanine and Orcein stainings, serum ceruloplasmin, serum copper, basal urine copper, and post-penicillamine (PCT) were measured according to the protocol previously described (13,14). Free copper concentration was calculated as previously described (15) and Zn corrected for albumin (16).
Diagnostic Score Used in the Study
The Proposed WD diagnostic score (Table 1) was applied for the WD patients and non-WD (control group) patients and compared with the diagnostic score published by Ferenci et al (Table 2). The Proposed WD diagnostic score ranged from 0 to 2 instead of -1 to 4. For the diagnostic score published by Ferenci et al, WD was diagnosed if the total score was 4 or more.
The numeric data was displayed as median and range. The quantitative data were displayed as proportion and percentage. Fisher's exact test and unpaired t test were used for the comparison of the categorical and numeric data, respectively. Receiver operating characteristic (ROC) analysis was used to identify the upper normal value for the diagnostic cutoff point for liver copper content, urine copper content, and serum ceruloplasmin (CP). Diagnostic sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and accuracy were ascertained for both scoring systems. Correlation between Zn and revised WD index were analyzed by Pearson Correlation. Data were analyzed by SPSS IBM® SPSS® version 23.0.
This analysis was registered as an audit of clinical practice.
There were 43 WD children with genetic mutation confirmed between 2005 and 2015. The median age at diagnosis was 11.2 years (1.6–17.3) and 53% were female. Of these, 28 were WD-non-ALF and presented with incidental findings of abnormal liver function test (n = 9), acute and chronic hepatitis (n = 7), neurological problems (n = 2), or as asymptomatic siblings (n = 10). Fifteen of the 43 were WD-ALF and presented with jaundice (n = 6), abdominal distension (n = 5), pedal edema (n = 3), and hematemesis (n = 1). Thirteen of the WD-ALF had chronic liver stigmata either ascites or splenomegaly and all of 11 WD-ALF who had liver biopsy performed had liver histopathology consistent with liver cirrhosis.
The results from various laboratory tests for WD are shown in Table 3. Comparing WD-ALF with WD-non-ALF, median values of serum alkaline phosphatase (ALP 92 [19–1120] and 298 [28–915] IU/L, P < 0.05) and Zn (5.8 [4.1–8.3] and 13.5 [6.1–22.2] μmol/L, P < 0.001) were significantly lower whereas the copper (14.5 [4.5–36.3] and 2.3 [2.11–14.6] μmol/L, P < 0.001) and free copper (7.92 [3.08–31.58] and 1.09 [0.02–6.01] μmol/L, P < 0.001) were significantly higher.
Because serum Zn level can be lower in acute stress such as ALF, we next carried out a critical analysis of liver function test including serum Zn, corrected serum Zn, and ALP in WD-ALF comparing to ALF from indeterminate causes. There were significantly lower level of corrected serum Zn (5.8 [4.1–8.3] and 9.8 [7.0–12.1] μmol/L, P < 0.001) and ALP (92 [19–1120] and 345 [159–758] IU/L, P < 0.05) in patients with WD-ALF compared to indeterminate ALF but no significant difference of copper and free copper level between WD-ALF and indeterminate ALF (P > 0.05). Interestingly, there was a significant difference in the ratio of AST to ALT between patients with WD-ALF, patients with indeterminate ALF (Table 3) and ALF from other causes (AST to ALT 2.83 [0.68–12.36], 1.05 [0.74–1.53] and 1.37 [0.18–7.3], P < 0.001 and P
< 0.05, respectively) (Supplementary Table 1, Supplemental Digital Content, http://links.lww.com/AA/C400). None of the patients with ALF of indeterminate etiology had ratios of AST to ALT more than 2. The triad of low Zn as well as ALP and AST/ALT ratio were useful in discriminating WD-ALF from indeterminate ALF.
Correlation of Zn and the Severity of Wilson disease Children
Serum Zn was significantly lower in WD-ALF patients compared with those with WD-non-ALF and ALF from indeterminate causes even when we corrected serum Zn for low albumin (Table 3). We used the revised WD index (12) to reflect the WD severity and compared the score to serum corrected Zn level in 26 WD children whose serum Zn data were available. There was a significant correlation between corrected serum Zn level and the disease severity score (r = −0.554, P = 0.004) (Figs. 1 and 2).
In order to revise the cutoff level of the 3 parameters in the diagnostic score published by Ferrenci et al mainly based on our data, ROC curves were assessed (Fig. 2). Two cutoff values for each parameter were selected, 1 cutoff value with a high sensitivity and 1 with high specificity to balance the probabilities of false-negative and false-positive cases when these cutoff values were applied in Proposed WD diagnostic score (Supplementary Table 2, Supplemental Digital Content, http://links.lww.com/AA/C400).
Liver copper content was high in patients with WD-ALF of which all patients had values more than 250 μg/g of dry weight liver whereas only 2 patients with WD-non-ALF had liver copper content lower than 50 μg/g of dry weight liver. The best cutoff value for high sensitivity and high specificity was 120 μg/g of dry weight liver as we have previously reported (17).
Basal urine copper content had the highest area under a ROC curve (0.939) compared with other parameters. Using the cutoff values of 0.6 and 2 μmol/d, 47 and 6 non-WD patients had negative and positive results, respectively.
Both cutoff values of 5 or 8 μmol/d in the urine copper content post-PCT had high sensitivity (83.8% and 81.1%, respectively) but the specificity of the cutoff of 5 μmol/d is better (40.5% vs 19.0%). We decided to use the cutoff value of 5 μmol/d for this study to detect positive results. In order to detect the negative results, we chose the cutoff value of 25 μmol/d, which has specificity of nearly 100% (99.5%) in our data and also as previously reported (6,11,12).
For the serum CP level, the original cutoff values of 0.1 and 0.2 g/L have the best sensitivity and specificity, respectively (18) in which only 2 WD-ALF patients had CP value of >0.2 g/L and none of the patients with AIH and NAFLD had CP value lower than 0.1 g/L.
Diagnostic Values of the Proposed Wilson disease Diagnostic Score
Based on the laboratory parameters that we identified, we developed the Proposed WD diagnostic score (Table 1) by excluding the genetic ATP7B analysis, adding the new parameters of the Triad (Zn, ALP, and the AST/ALT ratio) and changing the cutoff values for the urine copper content. Because Orcein staining is routinely used for histopathology tests in our center for all liver cases (6), we calculated the diagnostic value of Orcein staining for WD diagnosis using AIH and NAFLD as a control group (data not shown). To detect WD in children, Orcein staining had fair sensitivity, specificity, PPV, and NPV of 72.90%, 80.17%, 52.94%, and 90.65%, respectively. As a result, we also included the positive Orcein staining to gain 1 mark in the Proposed WD diagnostic score if quantitative liver copper measurement is not available.
The diagnostic values of the diagnostic score by the Proposed WD diagnostic score were calculated. The sensitivity, specificity, PPV, NPV, and accuracy were 87.0% (95% CI 73.7–95.1), 99.2% (95% CI 95.5–99.9), 97.6% (95% CI 87.1–99.9), 95.3% (95% CI 90.0–98.3), and 97.7% (95% CI 94.1–99.4), respectively.
When the diagnostic score published by Ferenci et al (genetics excluded) was used, 9 patients with WD had false-negative results and 11 patients with AIH had false-positive results whereas when we used our Proposed WD diagnostic score, 6 patients with WD had false-negative result and 1 patient with AIH had false-positive results (Supplementary Table 3, Supplemental Digital Content, http://links.lww.com/AA/C400).
In this study we show that serum Zn is significantly lower in our patients with WD who presented with ALF. In our patients with WD, regardless of presentation, Zn levels show a negative correlation with the disease severity as assessed by the revised WD index. One question is, does this reflect “real” or “apparent” Zn deficiency? Zinc can appear low in conditions of stress, systemic inflammation (19), and low albumin (16). We therefore tried to correct the Zn level with the method described (16) in an experimental rat study. The corrected Zn level in the WD-ALF patients when compared with those with WD-non-ALF, other mimic liver diseases or ALF of indeterminate cause was significantly lower.
Serum Zn may decrease as a result of its tissue redistribution in critically ill children (19,20). Zinc deficiency also exacerbates the inflammatory response (21,22). Zn supplementation in patients, for example, with hepatitis C-related chronic liver disease has been shown to have an anti-inflammatory effect (23). In the present study, both Zn and corrected Zn in WD-ALF was significantly lower than in then indeterminate cause of ALF signifying that this was not just a reflection of systemic inflammation. Zn deficiency has been associated with liver disease (24), particularly when it is accompanied with cirrhosis (25), but also with poor outcome of chronic liver diseases (23,25,26–28) and has been implicated in the pathogenesis of cirrhosis (25). Certainly, all our children with WD-ALF at presentation with histology were found to be cirrhotic (100%, N = 11).
A recent meta-analysis (29) showed significantly low serum Zn in autoimmune diseases and Pereira et al (30) demonstrated low Zn blood levels despite good nutritional status in children with AIH compared with healthy controls. The high prevalence of Zn deficiency in their study may reflect a higher significance of Zn deficiency in AIH in comparison to other chronic liver diseases. We therefore decided not to mark for low serum Zn in the Proposed WD diagnostic score, if there was a normal ALP and an AST to ALT ratio >2 to avoid false-positive diagnoses. Low serum Zn in isolation is not pathognomonic of WD-ALF, but it can alert the physician toward diagnosing a chronic liver disease such as WD and AIH.
Reports of Zn levels in patients with WD are few (1,31,32). One study reporting on 18 children with WD with liver and neurological manifestations demonstrated significantly lower levels of zinc compared with healthy controls (31). Iorio et al (32) hint at a possible link between zinc levels and severity of WD when they mention that in their series of 14 patients with mild disease Zn levels were normal. Whether low Zn is characteristic of a more severe WD-related hepatopathy should be further studied. Whereas Zn has been used in the treatment of WD (33) its potential role in the pathogenesis of this disease is less well understood. Some suggestions may come from work looking into how Zn metabolism is affected in chronic liver disease. The disturbance in copper metabolism seen in WD results in oxidative stress in the liver (34,35). Oxidative stress in the liver, as studied in alcoholic liver disease, has been shown to impair hepatic Zn homeostasis by affecting Zn transporters either directly or by altering their transcription levels, but also potentially by affecting Zn association with Zn proteins or by affecting transcription factors (36). Treatment with Zn in WD has been shown to improve these oxidative parameters (37). Zn supplementation has been shown to increase metallothionein induction in the gut and in this way reduces copper absorption by the body (38). In a similar but retroverted way, the competitive inhibition of Zn absorption by copper is a theoretical possibility for the low serum Zn in WD. There is some experimental evidence that suggests an increase in zinc content of the liver in the rodent model of Wilson disease (39) and something similar has been described in the human brain of patients with WD. In the liver, it has been speculated that this is due to zinc sequestration by the metallothioneins, which are in turn induced by the excess liver copper (40,41). No studies to our knowledge have looked at the relationship between liver tissue zinc content and serum zinc levels in WD. The negative correlation we have demonstrated between the revised WD index and the low serum zinc level, suggests that zinc may not only be an additional parameter to help in diagnosing WD, but also a surrogate marker of WD severity. Of course, further studies are warranted to investigate the importance of Zn in WD pathogenesis.
ALP also correlates with Zn deficiency. Low levels of ALP in patients with WD-ALF may reflect Zn deficiency, but the levels of ALP can also be affected by the protein concentration of the enzyme related to the liver function and bone turnover. The levels of ALP can influence the enzyme activity correlated with the Zn and Mg atoms (16,42). Normally, children have a higher ALP value than adult due to higher bone turn over, consequently, we interpreted the values for ALP with caution by adjusting for age according to the previous study (43). We had divided the children into 2 groups based on their ages and found that this helped ascertain the best cutoff values for the Proposed WD diagnostic score. However, 1 limitation of our cohort was that we did not have the complete data of Mg in our WD patients. Low bone and liver-isoform ALP has been associated with low Mg (44,45).
In addition, our WD-ALF children had a mild increase of ALT compared with the degree of hepatic impairment most likely due to the liver being cirrhotic. AST was higher than ALT, which may reflect the mitochondrial injury in WD pathogenesis. The importance of the AST/ALT ratio in diagnosing WD-ALF has been reported in adults (46–48). The Triad of Zn, ALP and the AST/ALT ratio (Fig. 2 and Supplementary Table 2, Supplemental Digital Content, http://links.lww.com/AA/C400) can further improve the diagnostic accuracy of WD. The Proposed WD diagnostic score can overcome the problems of WD-ALF diagnosis without waiting for the ATP7B analysis. Furthermore, Zn is the new parameter that can be used to diagnose and assess the severity in WD in children.
The present study suggests the usefulness of serum Zn as a new surrogate marker, not only to help diagnose WD, but also assess WD severity. Combined parameters of low serum Zn as well as ALP and AST/ALT ratio can discriminate WD-ALF from ALF from other etiologies. The Proposed WD diagnostic score appears to have high specificity and sensitivity. This proposed diagnostic score can be used for WD diagnosis while genetic results are awaited.
1. Van Biervliet S, Kury S, De Bruyne R, et al. Clinical zinc
deficiency as early presentation of Wilson disease. J Pediatr Gastroenterol Nutr
2. Basu A, Thanapal S, Sood M, et al. Catatonia: an unusual manifestation of Wilson's disease. J Neuropsychiatry Clin Neurosci
3. Jung KH, Ahn TB, Jeon BS. Wilson disease with an initial manifestation of polyneuropathy. Arch Neurol
4. Thomas GR, Forbes JR, Roberts EA, et al. The Wilson disease gene: spectrum of mutations and their consequences. Nat Genet
5. Ferenci P, Caca K, Loudianos G, et al. Diagnosis and phenotypic classification of Wilson disease. Liver Int
6. Koppikar S, Dhawan A. Evaluation of the scoring system for the diagnosis of Wilson's disease in children. Liver Int
7. Sturm E, Piersma FE, Tanner MS, et al. Controversies and variation in diagnosing and treating children with Wilson Disease: results of an international survey. J Pediatr Gastroenterol Nutr
8. Sintusek P, Chongsrisawat V, Poovorawan Y. Wilson's disease in Thai children between 2000 and 2012 at King Chulalongkorn Memorial Hospital. J Med Assoc Thai
10. Huster D. Wilson disease. Best Pract Res Clin Gastroenterol
11. Squires RH Jr, Shneider BL, Bucuvalas J, et al. Acute liver failure in children: the first 348 patients in the pediatric acute liver failure study group. J Pediatr
12. Dhawan A, Taylor RM, Cheeseman P, et al. Wilson's disease in children: 37-year experience and revised King's score for liver transplantation. Liver Transpl
13. Martins da Costa C, Baldwin D, Portmann B, et al. Value of urinary copper excretion after penicillamine challenge in the diagnosis of Wilson's disease. Hepatology
14. Muller T, Koppikar S, Taylor RM, et al. Re-evaluation of the penicillamine challenge test in the diagnosis of Wilson's disease in children. J Hepatol
15. Walshe JM. Clinical Investigations Standing Committee of the Association of Clinical Biochemists. Wilson's disease: the importance of measuring serum caeruloplasmin non-immunologically. Ann Clin Biochem
16. Naber TH, Baadenhuysen H, Jansen JB, et al. Serum alkaline phosphatase activity during zinc
deficiency and long-term inflammatory stress. Clin Chim Acta
17. Sintusek P, Dhawan A. Liver copper estimation, does liver biopsy size really matter? Hepatology
18. Roberts EA, Schilsky ML. American Association for Study of Liver Disease. Diagnosis and treatment of Wilson disease: an update. Hepatology
19. Dao DT, Anez-Bustillos L, Cho BS, et al. Assessment of Micronutrient Status in Critically Ill Children: Challenges and Opportunities. Nutrients
20. Negm FF, Soliman DR, Ahmed ES, et al. Assessment of serum zinc
, selenium, and prolactin concentrations in critically ill children. Pediatric health, medicine and therapeutics
21. Liu MJ, Bao S, Napolitano JR, et al. Zinc
regulates the acute phase response and serum amyloid A production in response to sepsis through JAK-STAT3 signaling. PLoS One
22. Wong CP, Rinaldi NA, Ho E. Zinc
deficiency enhanced inflammatory response by increasing immune cell activation and inducing IL6 promoter demethylation. Mol Nutr Food Res
23. Himoto T, Hosomi N, Nakai S, et al. Efficacy of zinc
administration in patients with hepatitis C virus-related chronic liver disease. Scand J Gastroenterol
24. Mohammad MK, Zhou Z, Cave M, et al. Zinc
and liver disease. Nutrition in clinical practice: official publication of the American Society for Parenteral and Enteral Nutrition
25. Grungreiff K, Reinhold D, Wedemeyer H. The role of zinc
in liver cirrhosis. Ann Hepatol
26. Goode HF, Kelleher J, Walker BE. Relation between zinc
status and hepatic functional reserve in patients with liver disease. Gut
27. Fota-Markowska H, Przybyla A, Borowicz I, et al. Serum zinc
(Zn) level dynamics in blood serum of patients with acute viral hepatitis B and early recovery period. Ann Univ Mariae Curie Sklodowska Med
28. Bartholomay AF, Robin ED, Vallee RL, et al. Zinc
metabolism in hepatic dysfunction. I. Serum zinc
concentrations in Laennec's cirrhosis and their validation by sequential analysis. N Engl J Med
29. Sanna A, Firinu D, Zavattari P, et al. Zinc
Status and Autoimmunity: A Systematic Review and Meta-Analysis. Nutrients
30. Pereira TC, Saron ML, Carvalho WA, et al. Research on zinc
blood levels and nutritional status in adolescents with autoimmune hepatitis. Arq Gastroenterol
31. Geetha A, Jeyachristy SA, Selvamathy SM, et al. A study on the concentrations of serum zinc
, non-ceruloplasmin copper, reactive oxygen and nitrogen species in children with Wilson's disease. Clin Chim Acta
32. Iorio R, Ranucci G. Wilson disease: a matter of copper, but also of zinc
. J Pediatr Gastroenterol Nutr
33. Socha P, Janczyk W, Dhawan A, et al. Wilson's Disease in Children: A Position Paper by the Hepatology Committee of the European Society for Paediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr
34. Nagasaka H, Inoue I, Inui A, et al. Relationship between oxidative stress and antioxidant systems in the liver of patients with Wilson disease: hepatic manifestation in Wilson disease as a consequence of augmented oxidative stress. Pediatr Res
35. Xu SQ, Zhu HY, Lin JG, et al. Copper ions stimulate the proliferation of hepatic stellate cells via oxygen stress in vitro. J Huazhong Univ Sci Technolog Med Sci
36. Sun Q, Li Q, Zhong W, et al. Dysregulation of hepatic zinc
transporters in a mouse model of alcoholic liver disease. Am J Physiol Gastrointest Liver Physiol
37. Guillard O, Piriou A, Gombert J, et al. Diurnal variations of zinc
, copper and magnesium in the serum of normal fasting adults. Biomedicine, 1979. 31 (7): 193–4.
38. Yuzbasiyan-Gurkan V, Grider A, Nostrant T, et al. Treatment of Wilson's disease with zinc
: X. Intestinal metallothionein induction. J Lab Clin Med
39. Allen KJ, Buck NE, Cheah DM, et al. Chronological changes in tissue copper, zinc and iron in the toxic milk mouse and effects of copper loading
40. Koropatnick J, Cherian MG. A mutant mouse (tx) with increased hepatic metallothionein stability and accumulation
41. Mercer JF, Grimes A, Rauch H. Hepatic metallothionein gene expression in toxic milk mice
42. Lowe D, John S. Alkaline phosphatase, StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2017.
43. Zierk J, Arzideh F, Rechenauer T, et al. Age- and sex-specific dynamics in 22 hematologic and biochemical analytes from birth to adolescence. Clin Chem
44. Nanji AA. Decreased serum alkaline phosphatase activity in hypothyroidism: possible relationship to low serum zinc
and magnesium. Clinical chemistry
45. Davidson JA, Urban TT, Baird C, et al. Alkaline Phosphatase in Infant Cardiopulmonary Bypass: Kinetics and Relationship to Organ Injury and Major Cardiovascular Events. The Journal of pediatrics
2017; 190: 49-55 e2.
46. Berman DH, Leventhal RI, Gavaler JS, et al. Clinical differentiation of fulminant Wilsonian hepatitis from other causes of hepatic failure. Gastroenterology
47. Korman JD, Volenberg I, Balko J, et al. Screening for Wilson disease in acute liver failure: a comparison of currently available diagnostic tests. Hepatology
48. O’Brien A, Williams R. Rapid diagnosis of Wilson disease in acute liver failure: no more waiting for the ceruloplasmin level? Hepatology