Associations between postoperative anaemia and specific drivers of unplanned readmission

Unplanned readmissions after major surgery place a substantial burden on healthcare systems, highlighting the need to identify their causes and develop targeted prevention strategies. We previously identified the most common reasons for readmission after major surgery, with infection and cardiovascular complications the two leading causes [1]. In addition, we found that a lower discharge haemoglobin concentration was associated with an increased risk of readmission [1]. However, we did not explore if there was an association between these two results. Given known associations between anaemia and worse outcomes in conditions like heart failure [2], we aimed to explore whether postoperative anaemia is associated with the most common causes of readmission after major surgery.

We reanalysed our previously described database of 14,632 patients who had undergone major surgery with adjudicated readmissions [1]. We aimed to determine if there was an association between discharge haemoglobin concentration and each of the top five categories of readmission identified in our previous study: infection; cardiovascular; gastrointestinal; respiratory; and musculoskeletal (online Supporting Information Table S1). Ethics approval was granted with a waiver of informed consent from the Austin Health Human Research Ethics Committee.

For each readmission category, we applied univariable and multivariable binomial logistic regression models. Discharge haemoglobin concentration was maintained as a linear independent variable for ease of interpretation. Covariates were those previously selected a priori. Restricted cubic splines were used for the other continuous covariates. The number of knots was selected between 3 and 6 using an iterative process to minimise Akaike information criterion for each of the multivariable models.

We found that a decrease in discharge haemoglobin concentration was associated with increased odds of readmission due to infection (p = 0.002) and cardiovascular complications (p = 0.011) (Table 1). No significant relationship was identified with readmissions due to gastrointestinal, respiratory and musculoskeletal causes.

For every 10 g.l^-1 decrease in discharge haemoglobin concentration, we found a 12% increase in the odds of readmission due to infection. From our previous study, we found that most unplanned readmissions driven by infection are due to wound complications (232/2048, 11%) [1]. Deep surgical site infections (60/2048, 3%) and cellulitis (29/2048, 1%) were the next leading causes of infectious readmissions. Similarly, we found a 12% increase in the odds of cardiovascular causes of readmission for every 10 g.l^-1 decrease in discharge haemoglobin concentration. The leading cause of cardiovascular readmissions identified in our previous study was heart failure (107/2048, 5%), followed by arrhythmia (69/2048, 3%) and thromboembolism (59/2048, 3%) [1].

Unplanned hospital readmissions after major surgery impose a significant burden on healthcare systems and are an urgent, unmet clinical need. Interventions designed to treat postoperative anaemia and, thereby, reduce complications should be targeted to the population most likely to benefit.

Patients who were iron replete had a significantly lower rate of surgical site infections compared with those who were iron deficient [3]. However, when the authors examined other common types of postoperative infection, such as urinary and respiratory tract, they found no significant differences [3]. Similarly, there are extensive data on the relationship between iron deficiency and heart failure. Several studies report a reduction in hospitalisation due to heart failure in patients who received an iron infusion compared with those who received placebo [4, 5]. Whilst there are significant data on the association between iron deficiency, anaemia and hospitalisation for numerous conditions, there is limited evidence on whether this relationship translates to postoperative patients. Our analysis shows that a lower haemoglobin concentration at the time of discharge following major surgery is associated with an increased risk of readmission due to infection or cardiovascular causes. We did not identify a significant relationship between postoperative haemoglobin and gastrointestinal, respiratory or musculoskeletal causes of readmission, meaning that these readmission categories should not be used as an endpoint for any study investigating the effect of postoperative anaemia management.

We acknowledge some limitations in addition to those identified previously [1]. This analysis examines the causes of readmission in system-based categories. It is, therefore, unclear if postoperative anaemia has a stronger association with specific diagnoses within that category. For example, while odds of readmission due to heart failure may be significantly higher than thromboembolism, we elected not to explore specific diagnoses due to the risk of multiplicity inherent in an exploratory re-analysis.

Based on these results, a potential interventional trial could be performed in patients with postoperative anaemia, examining whether the treatment leads to reduced incidence of readmission due to infection or cardiovascular complications.