Judgements of quality of care in NAP7 : a clarification from the authors

It is a pleasure to see another editorial [1] accompanying the 7th National Audit Project (NAP7) paper regarding obstetric peri-operative cardiac arrest [2]. We would like to raise one point to correct a possible misinterpretation in the editorial and clarify what we laid out in the NAP7 methods and results [3, 4] regarding causes of cardiac arrest and inferences about quality of care. Dr Monks et al. state, “anaesthetic management factors were considered the key cause of the majority of maternal cardiac arrests observed in NAP7” and further that “although maternal cardiac arrest is rare in the UK, suboptimal anaesthetic care contributes to a worryingly large proportion of cases”. However, in obstetric cases and throughout NAP7, key causes of cardiac arrest should not be misconstrued as representing poor care. First, cases could have multiple key causes, so in those where anaesthesia was cited, it was ‘a key cause’ not ‘the key cause’. Among all cases in NAP7 the ‘key cause’ included the patient in 82% of cases, anaesthesia in 40% and surgery in 35% [4]. Second, being a key cause does not equate to poor care. As an example, if a patient has a cardiac arrest due to anaphylaxis to atracurium, the key causes are the patient (due to allergic susceptibility to the drug) and anaesthesia (due to the fact that anaesthetic drug administration caused it). There is no blame or poor care implied. The case mix will also alter the proportions of each key cause. In the case of obstetric patients receiving anaesthesia care, the ‘patients’ are exclusively young, generally fit and well before labour and many (including the roughly half of cases whose only anaesthetic intervention is labour analgesia) are having no surgery. Consequently, anaesthesia will inevitably be a relatively higher causal factor in peri-operative cardiac arrests than the patient or surgery when compared with many other specialities. A comparison is vascular surgery where almost all patients have comorbidities, and all are undergoing surgery with inherent risk [5]. The data bear this out: in obstetric cases, key causes were the patient in 54%, anaesthesia in 68% and surgery in 29%, whereas in the vascular cohort, key causes were the patient in 88%, anaesthesia in 33% and surgery in 30%. Throughout NAP7 we estimated and reported quality of care quite separately from key causes [3, 4], recording it as ‘good’, ‘good and poor’, ‘poor’ or ‘uncertain’ with assessments made for the periods before, during and after cardiac arrest and overall. In general, poor care occurred before cardiac arrest with 92 (79%) of 117 ratings of poor care occurring in the phase before cardiac arrest [4]. In obstetric cases of cardiac arrest, care before cardiac arrest was rated poor in 18% with elements of poor care in 25% [1], compared with vascular surgery (10% and 34%, respectively) [5] and in all cases (10% and 31%, respectively). Overall care in obstetric patients was rated good in 53% and poor in none.