|Home | About | Journals | Submit | Contact Us | Français|
A previously well 5-year-old girl presented with new onset type 1 diabetes mellitus and diabetic ketoacidosis, and was found to be profoundly hyperlipidaemic. Further investigations showed that she had associated hypothyroidism. She responded to insulin and L-thyroxine treatments and her lipid profile returned to normal 2 months after diagnosis. Despite starting anticoagulant therapy early, she developed deep vein thrombosis of the lower limb. Her family screen did not demonstrate familial hyperlipidaemia or hypothyroidism. We discuss our patient’s diagnosis and management and highlight the challenges faced in this case. A lipid profile may be warranted in cases of concomitant diabetes and hypothyroidism to detect this problem and institute early treatment(s); monitoring for potential complications is warranted.
To our knowledge, this is the first case with association of diabetes, hypothyroidism and significant hyperlipidaemia to be reported in the literature. There are several important aspects highlighted by this case that led to the decision to report it. First, the profound degree of hyperlipidaemia represented a management challenge and consideration was given to starting plasmapheresis, but as the patient’s lipid profile was improving this was withheld. Second, it highlights the importance of insulin as a lipid as well as glucose regulator and the potential cross talk between insulin and thyroid hormone in maintaining lipid homeostasis. Thirdly, we wanted to highlight the risk of hyperviscosity due to dehydration and hyperlipidaemia, and the need for central access requirements to be weighed carefully against the risk of thrombosis in this case.
A previously well 5-year-old girl presented with 1 week history of tiredness, lethargy and pallor. She had no history of polyuria, polydipsia, polyphagia, nocturia, nocturnal drinking, enuresis, weight loss or recent infection. There was no family history of diabetes, autoimmune disease or dyslipidaemias.
On examination, she had pale waxy skin with cool peripheries. Her Glasgow Coma Scale was 15/15 and she was 10% dehydrated. She was tachycardiac (pulse rate 160 beats/min), and tachypnoeic (respiratory rate 44 breaths/minute) with Kussmaul breathing. Blood pressure was 120/68 mm Hg. Her height was below the 3rd centile with a midparental height on the 50th centile. Her weight was on the 50th centile. Fundoscopic examination showed clear optic disc margins. Abdominal examination revealed generalised tenderness on deep palpation and bowel sounds were present. She had no goitre and there was no evident focus of sepsis. On obtaining vascular access, the blood samples appeared notably lipaemic (fig 1).
Initial investigations revealed hyperglycaemia (blood glucose 29.8 mmol/l) and ketonaemia (blood ketones 4 mmol/l), and urine dipstick test revealed glycosuria and ketonuria. Arterial blood gas showed metabolic acidosis with partial compensatory respiratory alkalosis (pH 7.1, Po2 15.42 mEq/l, Pco2 3.2 mEq/l, HCO3 11mmol/l, base excess −18). HbA1c was elevated (19.2%) and while islet cell antibodies were negative, her GAD (glutamic acid decarboxylase) antibodies were elevated.
Lipid profile analysis showed elevated cholesterol of >1158 mg/dl (>30 mmol/l, normal range <4.4 mmol/l) and triglycerides >8850 mg/dl (>100 mmol/l, normal range 0.23–1.7mmol/l). Her plasma viscosity was raised (2.09, normal range 1.5–1.7).
Thyroid function tests showed elevated thyroid stimulating hormone (TSH) 400 mIU/ml and free T4 was not detectable; she also had positive thyroid peroxidase antibodies (2012 IU/ml, normal range 0–50 IU/ml). Quantitative assay of serum lipids revealed a 65% drop in serum lipid values by day 2, so plasmapheresis to remove excess lipids from the circulation was withheld. As her blood samples were lipaemic, it was difficult to obtain accurate electrolyte measurements on the lab analyser. Electrolyte measurements on blood gas analyser were used for monitoring as these measurements were unaffected by hyperlipidaemia. Her serum amylase concentration was normal and an abdominal ultrasound showed increased liver echogenicity with no evidence of pancreatitis.
As part of initial resuscitation, she received two 10 ml/kg boluses of 0.9% sodium chloride (NaCl) and then was started on 0.9% NaCl infusion with added potassium chloride; her fluids were changed to 5% dextrose/0.9% NaCl once blood glucose dropped below 14 mmol/l. In addition, an insulin (Actrapid) infusion was started at 0.1 U/kg/h to maintain blood glucose concentrations at a target of 4–10 mmol/l. She was admitted to the intensive care unit (ICU) for further management and radial and femoral lines were placed to allow access and monitoring of haemodynamic status. Given the risk of thrombosis associated with the pronounced hyperlipidaemia and increased plasma viscosity, she was started on tinzaparin (Innohep) prophylaxis 50 IU/kg/day. Once the diagnosis of hypothyroidism was confirmed, she was commenced on L-thyroxine 75 μg daily. On day 8, she complained of pain and swelling of her right leg. Deep vein thrombosis (DVT) of the right femoral and internal iliac veins was detected by Doppler ultrasound scan. She received tinzaparin 175 IU/kg/day for 3 months and the thrombus resolved with follow-up scans.
The metabolic acidosis corrected within 24 h while blood ketones and glucose were improving and returned to normal by day 2 and 3 of admission, respectively. She was switched to twice daily subcutaneous insulin regimen on day 3 of admission. Her cholesterol and triglyceride values continued to drop and returned to normal 2 months after the diagnosis and remained normal afterwards. At her clinic visit 1 year after this episode, she showed catch-up growth with both height and weight on the 50th centile and was developmentally normal. Both parents were screened for familial hyperlipidaemia, diabetes, and hypothyroidism and were normal. Unfortunately, she was lost to follow-up subsequently.
This 5-year-old girl presented with new onset type 1 diabetes mellitus, diabetic ketoacidosis (DKA), and hypothyroidism, and this was associated with significant hyperlipidaemia. Her clinical course was complicated by DVT secondary to hyperlipidaemia, hyperviscosity and the presence of femoral line. Her lipid profile normalised 2 months following commencement of insulin and L-thyroxine therapy.
Lipid abnormalities are common in children with new onset diabetes mellitus with DKA,1 and treatment with insulin leads to its resolution over a few weeks.1–4 In addition, isolated hypothyroidism can be associated with hyperlipidaemia.5 The severity of hyperlipidaemia seen in our patient is rare and is likely due to the simultaneous presence of diabetes and hypothyroidism of uncertain duration. As regards to the potential mechanisms involved in causing such a significant degree of hyperlipidaemia, it has been proposed that insulin deficiency, alone or in association with LPL gene mutations, results in impaired lipoprotein lipase action in adipose tissue and reduced lipid clearance.1,6,7
Regarding the management of severe hyperlipidaemia with DKA, plasmapheresis and other methods of lipid separation from plasma do have a role in management.8 Plasmapheresis removes excess lipids, improves metabolic acidosis, reduces plasma viscosity and the risk of pancreatitis and thrombosis. It was considered early in our patient but as the plasma lipid profile, metabolic acidosis, and clinical condition were improving it was withheld.
One of the significant complications that our patient had was DVT despite prophylactic anticoagulant use. Tinzaparin reduces the risk of thrombosis, releases lipoprotein lipase to the circulation, and reduces the concentration of circulating chylomicrones.5 However, it did not prevent DVT in this case, but this could be due to multiple factors including unknown duration of illness with dehydration, acidosis, hyperlipidaemia, hyperviscosity and central access. Early anticoagulant use is important in these cases and careful monitoring for the development of thrombosis is necessary to allow early intervention; benefits of central vascular access should be weighed carefully against potential risk of thrombosis. To our knowledge, this is the first reported case with profound hyperlipidaemia secondary to concomitant diabetes and hypothyroidism. We propose that a lipid profile needs to be considered when diabetes and hypothyroidism coexist to facilitate the detection and early intervention to correct hyperlipidaemia, and to avoid potential complications.
Competing interests: None.