Magnesium Sulphate for Severe Hand, Foot and Mouth Disease in Vietnam

Background:

Hand foot and mouth disease (HFMD) is a common infectious disease caused by a variety of enteroviruses. A small proportion of those infected with enterovirus 71 (EV71) develop neurological and systemic complications that may prove fatal. Over the past 15 years EV71 related HFMD has caused increasing epidemics of HFMD across Asia; in 2011, in excess of 100,000 Vietnamese children were diagnosed clinically with HFMD and 164 died. The neurological problem of most concern is brainstem encephalitis, causing autonomic nervous system (ANS) dysregulation that may progress rapidly to cardiopulmonary failure. Management of ANS dysregulation is difficult even in sophisticated western intensive care units. The phosphodiesterase-3 inhibitor, Milrinone, was reported to control hypertension and support myocardial function in a small informal study of severe HFMD compared to historical controls, but in practice treatment remains largely empirical. ANS dysregulation also occurs in severe tetanus and there is a body of evidence indicating that intravenous magnesium sulphate (Mg) is effective in controlling tetanus-associated cardiovascular instability. Mg is also used widely for eclampsia, severe asthma and pulmonary hypertension, and there are reports of rapid control of life-threatening autonomic hyperreflexia in patients with spinal lesions. Formal safety data in children are limited, but adverse effects appear to be infrequent. In a series of 24 severe EV71 confirmed HFMD cases managed recently at the Hospital for Tropical Diseases, Ho Chi Minh City, Mg was added when hypertension remained poorly controlled despite high-dose Milrinone. In all cases the blood pressure (BP) reduced within 30-60 minutes and remained stable subsequently on a continuous Mg infusion for 48-72 hours.

Hypothesis and Aims:

We hypothesize that early intervention with Mg, when ANS dysregulation first becomes apparent, will control cardiovascular instability and prevent progression to severe disease. The main aims are as follows:-

• To evaluate the effects of Mg on hypertension and ANS dysregulation in severe HFMD
• To evaluate the effects of Mg on outcome (death or neurological sequelae) in severe HFMD
• To evaluate changes in plasma/urine catecholamine levels, cardiac output and systemic vascular resistance (SRV) in severe HFMD, and to assess the impact of Mg on these parameters.

Study Design:

We plan a randomized double blind trial comparing intravenous Mg with placebo in children admitted to the pediatric intensive care unit with clinical HFMD, ANS dysregulation, and Stage 1 hypertension. Patients will be eligible for enrolment if the arterial blood pressure exceeds the 95th centile for age, gender and length/height for at least 30 minutes while the child is not distressed, they exhibit at least one other criterion for ANS dysregulation, and a parent/guardian gives written informed consent. Specific renal and cardio-respiratory exclusion criteria will apply.

A loading dose of 50mg/kg of either Mg or identical placebo will be given over 20 minutes followed by a maintenance infusion for 72 hours according to response, aiming for Mg levels 2-3 times normal in the treatment arm. All staff involved in clinical care will be blind to the treatment allocation, and Mg levels will be monitored and adjusted by independent doctors from another clinical facility.

The primary endpoint will be a composite endpoint of disease progression defined as occurrence of any of the following within 72 hours: specific blood pressure criteria necessitating addition of milrinone, need for ventilation, development of shock, or death. Secondary endpoints will include presence of neurological sequelae at discharge in survivors, and various measures of cardiac output and systemic vascular resistance, catecholamine and cytokine levels.

Based on 1:1 randomization, an anticipated relative reduction of the risk of progression of 50% (from 50% in the control arm to 25% in Mg recipients, information from our case series plus indirect evidence from studies on severe tetanus), 90% power and a two-sided 5% significance level, 85 patients per treatment group are required. To allow for some violations of our assumptions and losses to follow-up, we plan to recruit 190 patients (95 patients per treatment arm) into the study.

Potential Impact:

Given the very large numbers of HFMD cases being seen in hospitals in Asia currently, if Mg is shown to be effective in controlling ANS dysregulation and preventing severe HFMD complications this would be of major importance for paediatric care throughout the Asian region. Mg is cheap, readily available and safe whereas milrinone is expensive, has a significant side effect profile and is often not available outside major centres.

Inclusion criteria

• Age 6 months to 15 years
• Clinical suspicion of HFMD requiring PICU/HDU admission
• Considered severe enough to warrant invasive blood pressure monitoring by PICU/HDU staff
• Development of hypertension defined as follows:

• For children aged 1 year and over, at least 3 consecutive systolic blood pressure recordings above the 95th centile for age, gender and length (USA guidelines for defining Stage 1 hypertension in children, (Appendix 2)) measured invasively over a period of 20 minutes provided the child is not distressed or crying [30, 31].
• For children aged 6 months to 1 year, systolic BP > 100 mm Hg measured invasively on at least 3 occasions over a period for 20 minutes provided the child is not distressed or crying
• Plus one or more of the following criteria:

• Tachypnoea for age
• Irregular or labored breathing, but with SpO2 above 92% in air and normal ABG (pH, pCO2, pO2, HCO3 all within the normal range for the local laboratory)
• Resting heart rate > 150 bpm
• Mottled skin
• Profuse sweating
• Refractory fever
• Hyperglycemia
• Informed consent

Exclusion criteria

• Past history of hypertension, chronic renal, cardiac or pulmonary disease, or any neurological disorder
• Hypertensive emergency
• Already commenced milrinone or any other inotropic agents
• Respiratory distress with SpO2<92% in air or PaCO2>45 mm Hg
• AV block or any arrhythmia
• Acute renal failure