1 Essex, UK and 2 London, UK
EditorI was interested in the abstracts from the Lane Fox Respiratory Unit regarding their results in weaning and survival after prolonged ventilation.1 2 Following the recent NHS Modernisation Agency Report Weaning and long-term ventilation,3 the experiences of units like Lane Fox will become increasingly important.
Unfortunately, there are some problems with both the tables in the abstracts that make them difficult to understand. The first abstract stated that the total number of patients admitted was 153, of which 53 survived and 42 died (Table 17).1 What happened to the other 58? From the table in the subsequent abstract, it appears there were 111 survivors, so perhaps the 53 in Table 17 were the patients who were ventilator-dependent on discharge.2 Of the survivors, Table 17 appears to state 35% were ventilator-dependent on discharge, whereas the text states Of the survivors about one-half were ventilator-dependent at hospital discharge. Thus the 35% in the table presumably refers to the percentage of ventilator-dependent patients at discharge compared with the total number of patients admitted (53/153), with 53 of the 111 survivors being ventilator-dependent (48%). The text refers to ventilator dependency at discharge as 21/94 (22%) for in house ICU and 32/59 (54%) for outside ICU, but the denominators are the total numbers in each group admitted to the unit, not the total number of survivors at discharge, which would have been more meaningful in this context.
The second abstract is difficult to disentangle, mainly due to the inadequate and misleading row labelling of the table (Table 18).2 There is essentially a repetition of the same information in rows two and four. Row two, labelled All survivors (% of total patients), refers to all survivors to hospital discharge (% of all admissions). The final row four labelled Dead (% of total patients) is the number who died in hospital (% of all admissions), which is the same information as row two. However, the table omits the pertinent information about those who died between hospital discharge and 2 yr. Once you have worked out that row four does not give this information, you must start searching for itsome of it is in the text of this abstract, and some in the text of the previous abstract. The number of deaths between live hospital discharge and 2 yr appears to be a further 35 patients, derived from the admission to 2 yr survival of 50%, giving a post-live-discharge to 2 yr survival of 68%. This can be calculated to have included six deaths in the neuromuscular/chest wall group (post-live-discharge to 2 yr survival of 85%), 14 deaths in the COPD group (post-live-discharge to 2 yr survival quoted at 52%), and three deaths in the postoperative group (post-live-discharge to 2 yr survival of 82%). The text is less than helpful in stating postoperative patients had the highest mortality, neuromuscular/chest wall the lowest2 without making it clear this was in-hospital mortality. Additionally, it would have been interesting to know how many of those weaned at discharge remained off the ventilator at 2 yr.
While I appreciate abstracts must be concise, I found these interesting and important abstracts confusing, mainly because of the tables.
A. C. David
Essex, UK
EditorWe thank Dr David for her comments and are pleased that our abstracts were considered important. We apologize, however, that there was difficulty understanding the texts.
With reference to the first abstract,1 entitled Outcomes, cost and long-term survival of patients referred to regional centre for weaning from mechanical ventilation, the table has unfortunately been misprinted in the journal. The original table listed the number of survivors as 111, 53 of whom were using a ventilator at discharge from hospital. Thus, the overall percentage of patients who required ventilatory assistance at discharge was 35% (53/153) and over half of the survivors (58/111=52%) were free from any form of support at discharge. The text was written to emphasize this point. Unfortunately, the interquartile ranges have also been inaccurately reproduced. We have included a corrected table above (Table 1).
|
With reference to the second abstract,2 The influence of diagnosis on survival and success in weaning patients from prolonged mechanical ventilation, rows two and four of the table contain essentially the same information as it was thought this would help the reading of the abstract. In the original, the table was placed before the text and was intended to summarise the outcomes at hospital discharge. However, the long-term mortality data had to be abbreviated for the abstract. It was included as text because combining this information with the previous data on outcomes at hospital discharge would have added to the complexity of the table. In addition, follow-up data on those who remained ventilator-free at discharge were incomplete when the abstract was submitted. We agree that this is important.
We hope our comments answer Dr Davids concerns. Fuller data from our unit are being written up and we hope that they will be of interest when published.
D. V. Pilcher
A. C. Davidson
London, UK
References
1 Pilcher DV, Hamid S, Williams AJ, Davidson AC. Outcomes, cost, and long-term survival of patients referred to regional centre for weaning from mechanical ventilation. Br J Anaesth 2002; 89: 361P
2 Pilcher DV, Hamid S, Williams AJ, Davidson AC. The influence of diagnosis on survival and success in weaning patients from prolonged mechanical ventilation. Br J Anaesth 2002; 89: 362P
3 Critical Care Programme Report. Weaning and Long-term Ventilation. NHS Modernisation Agency, April 2002