Symptomatic hypercalcaemia in paediatric tuberculosis

To the Editors:

A 13-yr-old Congolese female presented in early spring with a 6-month history of productive cough, progressive exertional dyspnoea and weight loss of 4.5 kg. 2 weeks prior to presentation she had developed high fevers and increasing lethargy. She had immigrated to the UK at the age of 2 yrs. She had not been overseas since and there was no history of contact with anyone with tuberculosis (TB). She had no past medical history and was fully immunised except for bacilli Calmette–Guérin.

On examination she was cachectic and lethargic. There were no signs of clubbing, rickets or oedema. Her weight was 34 kg (less than third centile), height 148 cm (10th centile) and body mass index 15 kg·m⁻² (less than first centile). She was tachycardic and tachypnoeic with oxygen saturations of 90% in air. There was dullness to percussion and widespread coarse crackles on auscultation throughout the left lung field and at the right apex. The chest radiograph showed extensive left lung and right upper lobe consolidation (fig. 1). Her C-reactive protein was 160 mg·L⁻¹ and erythrocyte sedimentation rate was 90 mm·h⁻¹. Her corrected calcium and phosphate were low, 1.96 mmol·L⁻¹ (normal range 2.18–2.47 mmol·L⁻¹) and 0.80 mmol·L⁻¹ (0.75–1.50 mmol·L⁻¹), respectively, with albumin and alkaline phosphatase levels of 23 g·L⁻¹ (35–40 g·L⁻¹) and 118 mmol·L⁻¹ (30–200 mmol·L⁻¹), respectively. Her Mantoux test was strongly positive (20 mm induration) and acid-fast bacilli were present in her sputum, which subsequently cultured Mycobacterium tuberculosis. In addition to consolidation, her chest computed tomography scan showed cavities within the distal bronchi, widespread tree-in-bud appearances throughout the left lower lobe and extensive mediastinal, paratracheal and subcarinal adenopathy. Standard quadruple anti-tuberculous therapy (isoniazid, rifampicin, ethambutol and pyrazinamide) was started. In view of her malnutrition, nasogastric (NG) tube feeds were also commenced. Her serum 25-hydroxy-vitamin D₃ level, taken prior to starting anti-tuberculous therapy, was low at 20 nmol·L⁻¹ (normal range 75–200 nmol·L⁻¹), therefore supplementation with ergocalciferol (vitamin D₂) 10,000 IU·day⁻¹ was initiated. On discharge from hospital 20 days later her serum calcium was within the normal range.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Calcium and vitamin D metabolism. Continuous arrows represent movement of calcium and vitamin D and its metabolites. Dashed arrows represent the effects of vitamin D and hormones on calcium. Dotted arrows represent potential mechanisms of hypercalcaemia in tuberculosis. The chest radiography of our patient is included and shows extensive pulmonary consolidation and lack of subcutaneous tissue. PTH: parathyroid hormone.

1 week later she represented with nausea and vomiting, abdominal pain and aching muscles. Her respiratory symptoms had improved but she was continuing to lose weight and appeared very low in mood. Her serum calcium level was very high at 3.74 mmol·L⁻¹, with phosphate and albumin levels of 1.60 mmol·L⁻¹ and 33 g·L⁻¹, respectively. Urinary calcium was high at 7.82 mmol·L⁻¹ (1.87–3.39 mmol·L⁻¹), although her renal function tests were normal. The 1,25-dihydroxy-vitamin D₃ level was also high at 311 pmol·L⁻¹ (40–150 pmol·L⁻¹). Her parathyroid hormone was low at <0.7 pmol·L⁻¹ (1.1–6.9 pmol·L⁻¹) and 25-hydroxy-vitamin D₃ was normal at 144 nmol·L⁻¹ (75–200 nmol·L⁻¹). The ergocalciferol supplements, which she had been taking for 28 days, were stopped and NG feeding was restarted with reduced calcium. Her symptoms of hypercalcaemia improved rapidly but serum calcium remained high so she was started on intravenous fluids. Her serum calcium gradually decreased over several days. 3 months later she remains on anti-tuberculous treatment and the hypercalcaemia has not recurred.

The two sources of vitamin D in humans are metabolism in the skin in response to sunlight, and ingestion and absorption via the intestine (fig. 1). Vitamin D₃ (cholecalciferol; the most abundant form) is partially hydroxylated in the liver to form 25-hydroxy-vitamin D₃ (25(OH)D₃; calcidiol) which is then further hydroxylated in the kidney to form 1,25-dihydroxy-vitamin D₃ (1,25(OH)₂D₃; calcitriol). The latter is the active form of vitamin D₃ which causes increased gastrointestinal absorption of calcium, osteoclastic resorption from bone and renal tubular resorption, raising serum calcium levels. In addition, vitamin D inhibits parathyroid hormone which regulates serum calcium levels.

Hypercalcaemia has been described in adult patients with most granulomatous disorders, including TB, although the majority are asymptomatic. Rates of hypercalcaemia in TB in adults vary widely from 6 to 48% [1, 2]. Symptomatic hypercalcaemia in TB is less common, with rates of ∼3% in adults [3], although there are no data in children. Allowing for variation in methodology, differences between populations are probably due to sun exposure and diet, particularly the latter.

Attempts to predict which patients will develop hypercalcaemia have been unsuccessful as there is no clear relationship between serum calcium level and type or stage of disease: some patients are hypercalcaemic at diagnosis, while others develop hypercalcaemia while on treatment. In addition, there is discrepancy between studies correlating serum calcium levels and radiological extent of disease, with some finding an association and others finding no relationship.

There are only a few cases in the literature over the last 25 yrs which have reported children with hypercalcaemia associated with TB, often with severe disease (table 1) [4–10]. Our case describes a teenage female with extensive pulmonary TB associated with symptomatic hypercalcaemia that persisted despite cessation of vitamin D supplementation. It is the first report of symptomatic hypercalcaemia in a child on both anti-tuberculous therapy and vitamin D supplementation.

The mechanism underlying hypercalcaemia in TB is thought to be over-production of 1,25(OH)₂D₃ in extra-renal macrophages in granulomas, for example in the alveoli. It has been suggested that the production of 1,25(OH)₂D₃ has a protective effect against oxidative injuries from the nitric oxide burst from granulomatous macrophages [11]. This was initially studied in other granulomatous diseases, such as sarcoidosis. In TB, 1,25(OH)₂D₃ has been shown to be produced by alveolar macrophages from a patient with pulmonary TB [12], and high serum levels have been found in anephric patients, further supporting an extra-renal source [13]. Because production is extra-renal, the normal feedback control of 1,25(OH)₂D₃ production by parathyroid hormone is ineffective.

However, TB-associated hypercalcaemia has also been reported in patients with normal [4] and low [14] levels of serum 1,25(OH)₂D₃. Other causes may include an overdose of vitamin D or calcium supplements. An alternative possibility is abnormal activation of parathyroid hormone-related protein, which is the usual aetiological agent of hypercalcaemia of malignancy and is raised in some patients with sarcoidosis [15].

Recent studies have focussed on the association between low vitamin D levels and active TB [16]. However, despite investigation of vitamin D levels and supplementation becoming increasingly common in clinical practice, the mechanism is not fully understood. This case raises the possibility that extra-renal production of 1,25(OH)₂D₃ in TB may inappropriately lower serum 25(OH)D₃ resulting in, or at least contributing to, this finding.

It is likely that the hypercalcaemia in our patient was multifactorial. Normal 25(OH)D₃ and high 1,25(OH)₂D₃ indicate that the hypercalcaemia was secondary to granulomatous disease rather than vitamin D toxicity. However, the fact that our patient was not hypercalcaemic at presentation suggests that it may not have been due to disease alone. The vitamin D supplementation would have increased calcium absorption from the intestine at the same time as she was receiving highly nutritious NG feeds. It is likely that this, in combination with extra-renal 1,25(OH)₂D₃ production, explains her hypercalcaemic state.

The management of hypercalcaemia in TB involves treatment of the underlying cause, together with measures to reduce the serum calcium. Restriction of calcium and vitamin D supplements is sometimes all that is required, although intravenous fluids may be used for more rapid normalisation of serum levels. A variety of other treatments have been used: ketoconazole, hydroxychloroquine, bisphosphanates and corticosteroids [6, 8, 17]. The mechanisms for these therapeutic measures are not fully understood, although corticosteroids are believed to reduce the concentration of 1,25(OH)₂D₃ by inhibiting hydroxylation to this metabolite [16]. Unless absolutely necessary, polypharmacy should be avoided as patients already have a significant medication burden and additional medications may affect compliance. In our case supplement restriction and intravenous hydration was sufficient.

Our case illustrates the infrequently recognised and poorly understood phenomenon of hypercalcaemia in TB. Children with TB are increasingly being tested and treated for vitamin D deficiency. However, there are no clear guidelines for dosage and duration of supplementation, and further studies are needed to address this. It is therefore important to monitor serum calcium and vitamin D levels especially after starting concurrent supplementation and nutrition, and to have a high index of suspicion for hypercalcaemia when presented with gastrointestinal or musculoskeletal symptoms.

• ©ERS 2011