Background

Preoperative anaemia including iron deficiency anaemia (IDA) is a well-established perioperative risk factor, but unfortunately, most studies on iron therapy to IDA in abdominal surgery have been negative (Richards et al. 2023). Similar observations have been made in hip and knee arthroplasty (THA/TKA) (Scrimshire et al. 2020). However, these studies have almost uniformly been done without an Enhanced Recovery After Surgery (ERAS) setting which otherwise is well known to decrease postoperative morbidity (Kehlet 2020). More recently, a well-established implemented ERAS programme with a median length of stay (LOS) of 1 day after THA/TKA in a large series of consecutive patients has demonstrated preoperative anaemia to be a risk factor for postoperative complications using a machine learning setting (Jørgensen et al. 2024).

Main text

In September 2022, the multicentre fast-track hip and knee replacement collaboration (Jørgensen et al. 2024) initiated a further detailed prospective registration of patient comorbidities including iron status. Here, the preliminary data from the initial 3655 patients having surgery until July 2023 is reported and summarised in Table 1.

Table 1 Postoperative outcomes
Full size table

Again, these data within an effective established ERAS setting and complete follow-up clearly demonstrate that patients with IDA have more outcome problems with slightly longer LOS, an increased fraction of patients with LOS > 2 days, 30 days and 90 days readmissions (Table 1). However, the study´s detailed preoperative patient demographic assessments in the lower panel of Table 1 illustrate that in addition to the clear outcome differences, seven other demographics were different in IDA patients vs. the non-anaemic group: a slightly higher age, an increased proportion living alone, an increased proportion receiving home care, fewer having a job, an increased proportion using a walking aid, an increased proportion with two or more falls within 3 months preoperatively and a higher frequency of polypharmacy (> 5 types of drugs).

Unfortunately, all of these, except maybe preoperative polypharmacy and fall tendency, maybe non-modifiable within a perioperative setting. Furthermore, the causes of LOS > 2 days and 90 days readmissions were diverse and with a considerable number being related to pain, and surgical complications (Fig. 1). Thus, if planning a conventional all-patient inclusion RCT and with a “hard” composite outcome of LOS > 2 days and 30- to 90-days readmissions, the incidence would be 22.8% according to the presented data. If considering an 8% reduction to be achievable when providing preoperative i.v. iron treatment, 2 × 371 IDA-patients would be needed for a power of 80% and a significance level of 0.05. However, given the considerable amount of postoperative morbidity which may not be related to anaemia or iron deficiency and the additional non-modifiable factors/risk differences without any expected association with the effects of preoperative iron IDA correction, the degree of outcome improvement is dubious.

Fig. 1
figure 1

Pie charts of the distributions of causes of LOS > 2 days and 90 days readmissions in IDA and non-anaemic patients

Full size image

The two main questions to be answered are therefore how to improve outcomes in IDA patients, and how to measure the potential of iron therapy. Although it is obvious that the fraction of patients presenting with established preoperative IDA should be included in interventional trials, the most suitable trial design and outcome are less straightforward as the conditions for a conventional RCT seem uncertain. This is problematic and may not be easily solved by increasing the number of patients, as increasing the number of recruiting sites or countries may influence outcomes associated with non-protocolised aspects of care (e.g. ± ERAS protocols). Consequently, a classical randomised clinical trial with a composite or single primary “hard” outcome may not be a feasible way forward to study the potential benefits of preoperative iron therapy for patients with IDA. Instead, we may need to consider using hierarchical primary outcomes as seen in some recent studies within internal medicine (Gasparyan et al. 2022), in order to provide information on potential prioritised benefits throughout the peri- and early postoperative period.

Another possibility could be a primary exploratory observational study on complications or improvements with an established/ expected association with anaemia and iron depletion. Such observational studies may serve as a basis before considering a hierarchic or classical RCT. Other potential outcomes of interest include non-invasive VO2max estimation (Sørensen et al. 2020; Plumb et al. 2023) or patient-reported outcome measures (PROMS), despite their limited value compared to the objective assessment of recovery (Wainwright and Kehlet 2022).

Conclusion

This detailed study in a well-defined ERAS setting demonstrating worse postoperative outcomes in patients with IDA emphasises the need for detailed observational and mechanistic studies on the potential effect of preoperative iron treatment. This is due to the considerable burden of patient-related risk factors which may not be modifiable by correction of IDA alone. Such studies will be important for rethinking study design and outcomes prior to initiating extensive definitive randomised studies (Kehlet and Lobo 2024).