In Rwandan ACS patients undergoing surgery with an indication for postoperative ICU admission, more than half of patients were admitted to the ward. Of those admitted to the ICU, only 65% were admitted within 6 hours postoperatively. Lack of ICU access and delayed ICU access were associated with increased mortality and length of hospital stay. Increasing ICU capacity has the potential to improve patient mortality as well as reduce hospital length of stay, which has been associated with overall healthcare costs.
The timely post-operative ICU admission rate was 24.9%, comparable to 23.3% in a previous study in Uganda and 31.2% in a study in Brazil [7][26]. However, the immediate ICU admission rate was much lower than that reported from a study in China (91.9%)[13][27]. This discrepancy in the rate of timely ICU admission may be due to differences in ICU bed availability in different settings. Uganda has an ICU bed ratio of 0.13 beds per 100,000 population [5][25] and Brazil has 22 ICU beds per 100,000 populations [7][26] whereas China has a capacity of 4.8 ICU beds per 100,000 population.[13][27] Additional differences may be impacted by the fact that those studies included patients from all specialties while this study only evaluated a specific group of surgical patients.
Our study found a mortality rate of 48.1% in the group that lacked post-operative ICU access. This mortality was comparable to the mortality (43%) in a study done in Tunisia about the decision to refuse admission to the ICU[11]. In a study in western France,[10] there was a 30% mortality rate associated with ICU admission refusal. These disparities in mortality in patients lacking access to ICU could be due to differences in settings where critically ill patients can be provided improved medico-surgical care in absence of a specific critical care unit or service. In Rwanda, efforts have been implemented to improve postoperative care and monitoring in the surgical ward, with higher priority patients being positioned nearby to the nursing station for closer monitoring. Despite these efforts, amongst patients admitted to the ward, the mortality was 39%, suggesting additional opportunities for improvement like increasing number of nurses and resources for monitoring.
Patients with delayed ICU access had a higher mortality rate than those admitted to the surgical ward. We hypothesize that the healthcare providers appropriately identified these patients as needing ICU access, but there were no beds. Rather than admit these patients to a surgical ward with inadequate services, these patients delayed in the PACU. Other studies evaluating mortality in patients with delayed ICU admission carried out in Bangladesh, Brazil, and Chicago respectively that found mortality rates ranging from 30–51%.[6],[7],[12] These patients were recognized as needing ICU admission, hence they were not admitted to the surgical ward. These patients spent prolonged time in the PACU, which is a monitored environment, but does not have the level of staffing or support as a traditional ICU environment. Further details on the ICU cares received could help elucidate how many of these patients would benefit from expanded ICU bed access.
The mean length of hospital stay in the patients who were immediately admitted to the ICU was shorter compared to patients who did not have post-operative ICU access. This short hospital stay in patients admitted to ICU was comparable to previous study done in Pakistan which has revealed average hospital stay of 12 days.[28] Patients with delayed ICU admission in our study stayed an average of 7.1 days in hospital which was low compared to 13 days from Churpek et al.[12] and 19 days from Yun Su Sim, et al.[29] The short median days in overall stay in hospital within our study could be explained by the early death that has occurred in severely ill surgical patients with delayed ICU access. Differences in hospital length of stay may also differ based on the underlying diagnosis, operation performed, and other patient characteristics. Improving ICU access to decrease length of hospital stay and it’s associated costs has the potential to provide benefits to the individual patient as well as the overall healthcare system through more efficient use of resources and efficient patient care.
There were several limitations to this study. First, the study was limited to 2 hospitals, so that the study’s results cannot be universally applied to every hospital. Although, definition of patients who need post-operative critical care management was done using CARES surgical risk calculator prior to surgery, the decision for ICU admission was at physician discretion (including both ICU and surgeon discretion) and represents a possible source of bias. We did not collect data on daily ICU bed availability. We did not collect specific data on ICU needs or cares received, so we cannot make conclusions on how ICU cares may have influenced outcomes. We were unable to assess long-term prognosis (involving period of 6 months or 1 year) associated with post-operative ICU access. Other outcomes, such as quality of life at discharge and outside the hospital would be important outcomes in addition to in-hospital mortality. We were unable to evaluate factors linked to ICU access but we had strengths like the universal tool, CARES surgical risk calculator, to identify patients who will need post-operative critical care management.
In conclusion, there was a strong association between lack of post-operative ICU admission and mortality and hospital length of stay in critically ill ACS patients. Not only is early surgical intervention paramount in this patient population, but also appropriate and timely post-operative disposition are keys to avoid preventable death.