Long-term survival and mortality
Eight of the 12 papers reviewed identified inferior long-term survival or higher mortality rates for those who had prolonged LOS in ITU in comparison to ‘normal’ LOS or a control. Mahesh et al.(10) reported overall survival at one year for prolonged ICU stay to be 89.2% vs 97.8% for the control group of LOS of less than three days, and 81.2% vs 93.6% at three year follow-up. Heimrath et al. (22) reported mortality rates of 14% in the prolonged LOS group vs 4% for the control (p=0.001). Elfstrom et al. (23) stated the mortality risk of groups with LOS 3-7 days and 7-14 days to be 50% higher than that of the ‘normal’ LOS. High mortality rates for those with prolonged ITU LOS were also reflected in findings by Gaudino et al.(9) where 57 participants survived to discharge and 78.9% of them died during follow-up (over an average of six years), of which 29 deaths were attributed to cardiac causes. However, it is important to note that this particular study had a small sample size (n=57) with no control population for comparison, therefore limiting generalisability of their findings to the wider population. Hassanet al. (24) and Hein et al.(5) reported poorer long-term survival in those with prolonged ITU stay that was more marked in the first six months to one year post-discharge. Long-term survival was 70% in the prolonged group vs. 90% in the control (p= < 0.001) (24)with similar results produced by Hein et al.(5) with 66% survival in the prolonged group vs. 91% in the control (p=0.001). There was a steep decline in survival in the first six months of follow-up, but the six-month to three-year period was comparable (with a 7% difference in decline for the prolonged group vs. 4% for the control) which led the authors to conclude that chances of survival are comparable if they overcome the first six months post-operatively (5) . Similar findings and conclusions were made by Hassan et al. (24) who demonstrated inferior long-term survival amongst patients with prolonged ITU LOS, especially within the first year after discharge, stating that the greatest burden of mortality is within the first year post-operatively.
Conversely, Silberman et al. (25) found that a significant number of patients went on to experience reasonable long-term survival. Of those with LOS ranging from 1-2 days and 1-3 weeks, 90% were alive at one year follow-up. Eighty per cent of those with a LOS of one week and 50% of those with a LOS of two weeks were alive at five years. This was the only study to identify a proportional relationship between ITU LOS and overall late survival. They also found that for those with an extremely prolonged LOS in ITU (more than 30 days), the highest rate of mortality is in the first year. Manjiet al. (26) took a different approach to their analysis by looking at ‘functional survival’ which they defined as patients being alive and not institutionalised. They found that the control group achieved 94.9% functional survival at one year vs. 73.9% for prolonged ICU stay and 84.9% vs. 53.8% at five years, which was considered a successful outcome by the authors. However, although the majority were alive and non-institutionalised, it is not known what QoL was like for these patients or whether they were requiring care or support at home as no assessment of QoL of functional status was made.