Diagnosis and management after life threatening events in infants and young children who received cardiopulmonary resuscitation.

Introduction

An apparent life threatening event has been defined as "an episode that is frightening to the observer and that is characterised by some combination of apnea (central or occasionally obstructive), color change (usually cyanotic or pallid), marked change in muscle tone, choking, or gagging".[1] When mouth to mouth resuscitation is needed there is a particularly high risk of subsequent sudden death. In one survey 10 of 76 patients with such events died.[2] We have used physiological recordings in hospital and at home to identify mechanisms for these events and report our findings in a subgroup that received mouth to mouth resuscitation.

Patients and methods

Over 44 months 157 patients (96 boys) who had suffered one or more events out of hospital and received mouth to mouth resuscitation were referred for clinical management to our department at the Royal Brompton Hospital. One hundred and thirteen had been born at term and the others at a median gestational age of 32 weeks (range 24-36); none of the latter had been discharged from their neonatal unit on additional inspired oxygen. Median postnatal age at referral was 2.8 months (range 1 week to 96 months); 128 were 6 months and under and 18 were over 12 months of age. Forty six patients had a single event, 59 had two-five, 21 had six-10, and 31 had more than 10 events before referral.

Conditions considered as triggers for events, such as infections, anaemia, and biochemical disturbances, were excluded or treated. In some cases electroencephalography (EEG), standard electrocardiography, 24 hour electrocardiography, barium swallow, or oesophageal pH studies were undertaken but failed to identify abnormalities considered relevant to the events. In 19 a previous sibling had died suddenly and unexpectedly. In nine patients suffocation was suspected at referral.

Physiological recording in hospital

Long term (8 hours to 3 weeks) analogue tape recordings of arterial oxygen saturation ([Sao.sub.2]) (Nellcor N-200 modified to provide beat to beat data), the plethysmographic (pulse) waveforms (to validate [Sao.sub.2]), breathing movements (Graseby capsule), electrocardiograms, and skin (transcutaneous) [po.sub.2] with the sensor heated to 43 [degrees] C (Kontron 821S) were performed on 150 patients.[3] Recordings were printed on an ink jet chart recorder (Siemens Mingograf) at 3.1 mm per second.

Seven did not undergo recordings. Reasons were suspected deliberate suffocation (by a parent) in two, a single event in two older children (13 and 48 months), and lack of inpatient facilities in three. In one of the latter, death occurred after referral but before admission.

Physiological recordings at home

Thirty nine patients with recurrent events but normal recordings in hospital, 20 with abnormal recordings in hospital but no captured clinical events, and two who did not undergo hospital recordings underwent oxygen event recordings (Research Monitoring Systems, or Kontron) at home.[4] Date, time, [Sao.sub.2], plethysmographic waveforms, skin [po.sub.2], breathing movements, and electrocardiograms were continuously monitored and recorded on to a 128 kB digital memory card (86x54x3 mm), which was continually updated with the last 20-30 minutes of data. In addition a log of use of the event recorder was stored. Skin [po.sub.2] was the only variable to register an alarm, thereby identifying sudden hypoxaemic episodes to the parent.[5] After an event the memory card was transferred for analysis.

Analysis of data

Baseline [SaO.sub.2] was considered abnormally low if the value assessed from the recording was <95% (based on data from healthy term and preterm infants.[6 7]). Abnormal hypoxaemic episodes were considered …