In August 2004, a previously healthy 6-year-old female presented with pica and emotional lability. Over the following months she developed progressive behavioral and neurologic symptoms: She became withdrawn and less verbal with repetitive stuttered speech, and her balance, co-ordination, and fine motor skills declined. By November, she could no longer stand independently, tended to fall backward, and developed a high steppage “cock-like” gait.
Magnetic resonance imaging (MRI) indicated hyperintensity in the basal ganglia, suggesting Mn accumulation. Blood samples analyzed with high-resolution inductively coupled plasma mass spectrometry confirmed elevated levels of Mn in whole blood (39.7 μg/L; normal, 4.3–15.9 μg/L) and serum specimens (3.2 μg/L; normal, 0.3–1.0 μg/L). Severe Fe deficiency [ferritin, 5 μg/L (normal, 12–140 μg/L); serum Fe, 5 μmol/L (normal 5–28 μmol/L)] and polycythemia [red blood cells (RBCs), 7.6 × 1012/L; normal, 3.9–5.3 × 1012/L) were identified; polycythemia was attributed to elevated Co (3.9 μg/L; normal: 0.7–0.8 μg/L). Blood lead was normal (2.9 μg/dL; normal, 0–4.6 μg/dL) in October 2004 but elevated (17.6 μg/dL) in early August 2005. Investigations failed to determine a cause for the manganism. The patient’s karyotype was normal. A liver biopsy ruled out hepatic dysfunction; however, liver Mn was elevated (34 μg/g; normal, 0.22–4.6 μg/g). Normal blood copper levels ruled out Wilson’s disease.
Treatment included phlebotomies for the polycythemia (November 2004–February 2005, October–November 2005), ethylene-diamine tetraacetic acid chelation for the Mn overload (March–December 2005), and Fe therapy (November 2004–July 2005, October–November 2005). Significant improvements in gait, retropulsion, and motor skills that were observed after the initiation of the phlebotomies and chelation plateaued after a few treatments.
In July 2005, Fe supplementation was stopped; a precipitous drop in the patient’s Fe levels immediately followed. Her plasma and RBC Mn levels began rising again ( and ). By August her condition had deteriorated: Her pica returned, she fell frequently, and she needed assistance where she was previously independent. Neurologists estimated that one-fourth of her improvements were lost. Phlebotomies and oral Fe therapy were re-initiated in October.
Timeline of Mn levels in plasma for the patient and her family (immediate and extended), March–October 2005. Abbreviations: Max, maximum; Min, minimum.
Timeline of Mn levels in RBCs for the patient and her family (immediate and extended), March–October 2005. Abbreviations: Max, maximum; Min, minimum.
The patient resided with her mother, father, and 7-year-old sister in an urban center in New Brunswick (NB), Canada. Since 2000, they have spent summers at their nearby cottage—weekend visits in June, followed by full time residence in July and August.
Water testing was conducted; municipal water consumed at the primary residence had nondetectable Mn. At the cottage, a sandpoint well used between 2000 and 2003 had Mn concentrations of 1.7–2.4 mg/L (the health guideline is 0.5 mg/L) (World Health Organization 2006
). Spring water and a neighboring cottage well used in 2004 had non-detectable concentrations and 1.7–2.2 mg/L Mn, respectively. In 2005 municipal water was brought to the cottage for drinking, but use of well water for washing and cooking continued.
A food history revealed that the patient and her family consumed more vegetables than a typical Canadian family, particularly Mn-rich leafy green vegetables and pineapples, but they were not vegetarian. Foods consumed by the patient and her sibling were very similar with one exception: The sibling consumed soy milk due to lactose intolerance, whereas the patient consumed dairy. Both took half of a children’s vitamin supplement daily.
No inhalation exposures were identified. No industrial Mn releases were reported in the vicinity of either the primary residence or the cottage (Environment Canada 2003
). Neither residence was located in high traffic environments; excessive exposure to methyl-cyclopentadienyl manganese tricarbonyl was therefore unlikely. Parental occupations did not involve exposure to Mn.
Site visits to the primary residence and cottage did not identify excessive exposure sources unique to the patient. No recent renovations, painting, or household products containing Mn were identified. Household members did not participate in hobbies associated with Mn exposure such as glasswork or ceramics. The father infrequently used a welding torch in the detached garage of the primary residence; the children were restricted from the garage during welding but otherwise had free access. During the summer months, vegetables consumed by the patient and her family came predominantly from their cottage garden. Agricultural products containing Mn (e.g., fungicides) were not used. Satellite imagery and site visits did not identify other Mn sources in the vicinity of either residence.
Exposures were very similar for the whole family and virtually identical for the patient and her sister, as they spent the majority of their time together. Only pica was unique to the patient.
Assessment of other persons at risk
The patient’s parents and sibling were asymptomatic. Between March and June 2005, the immediate family and five extended family members (maternal grandparents, paternal grandfather, and the father’s two siblings in lieu of the deceased paternal grandmother), were tested for Mn. Results indicated elevated plasma Mn ranging from 1.9 to 2.8 μg/L (normal, 0.3–1.0 μg/L) for everyone except the patient’s sister; her plasma Mn was normal (0.6 μg/L) in March but elevated (2.3 μg/L) when retested in May. All those tested had RBC Mn within the normal range ( and ).
Genetic counseling did not identify a familial cause. Both maternal and paternal families are NB Acadian for at least three generations, with no identified consanguinity. No history of Fe deficiency, neurologic disorders (e.g., Parkinson disease), or metabolic disorders were identified.
Early stages of Mn toxicity include muscle tremors and can sometimes resemble Parkinson disease, which has been proposed as a sentinel among selected occupational groups (Harvard Medical School 2000
). We examined hospital separation data for Parkinsonism between 1991 and 1995 in which NB rates were similar to other Canadian provinces (Kontakos and Stokes 2000
Specialists considered this case unique. They queried an international metabolic list-serv to identify other clinically similar cases, and identified a girl with hypermanganesemia of unknown origin in a consanguineous family in the United Kingdom. Her MRI indicated increased intensity in the basal ganglia, and she had a “cock-like” gait. She was polycythemic with elevated iron-binding capacity but had normal serum Fe. We also identified a single case from the literature—a boy from California with polycythemia and manganism explained by liver dysfunction (Gospe et al. 2000