Timing of intravenous iron for treatment of anaemia in surgical patients: a systematic review and network meta-analysis

Summary

Background

Anaemia complicates recovery in surgical patients. Intravenous (IV) iron supplementation shows promise in improving outcomes, but optimal timing remains uncertain. In this review, we compare the efficacy, safety, tolerability, and outcomes between preoperative and postoperative IV iron supplementation.

Methods

In this systematic review and network meta-analysis, we searched PubMed, EMBASE, Cochrane Library, and Web of Science from inception to May 1, 2025, for randomised controlled trials (RCT) investigating IV iron supplementation in surgical patients either 7–30 days before surgery (preoperative) or 0–30 days after surgery (postoperative). Studies were excluded if they included patients with critical illness or prior transfusion or if iron was given outside the defined time frames or with other agents. Two reviewers independently appraised the data and extracted summary estimates from published reports. The primary outcomes were: (1) proportion of patients who received blood transfusion; (2) change between the baseline haemoglobin level and the haemoglobin level on postoperative day (POD) 7 and POD30. Data processing was conducted based on frequentist network meta-analysis. The risk of bias was assessed using the Cochrane Risk of Bias tool. The protocol is registered with PROSPERO, CRD42024533265.

Findings

Among 129 identified studies, 22 RCTs with 3026 patients were included. All included studies had a low (n = 6) or moderate (n = 16) risk of bias. Compared to controls, postoperative IV iron supplementation reduced transfusion rates (RR 0.80, 95% CI 0.68–0.94; I² = 0.0%). Postoperative IV iron supplementation did not affect haemoglobin levels (MD −4.51, 95% CI −9.75 to 0.72; I² = 90.3%) at POD7 but increased haemoglobin levels (MD 5.45, 95% CI 2.70–8.20; I² = 45.5%) at POD30. In comparison, preoperative IV iron supplementation resulted in higher haemoglobin levels than postoperative supplementation at POD30 (MD 6.67, 95% CI 1.61–11.72) but did not influence transfusion rates (RR 0.91, 95% CI 0.72–1.15; I² = 0.0%).

Interpretation

Our results suggest that postoperative IV iron supplementation reduces transfusion rates, while preoperative supplementation improves haemoglobin recovery. Clinicians may choose either strategy in an individualised, patient-centered manner. These conclusions should be interpreted with caution due to heterogeneity among included studies, limited data for subgroup analyses, and the absence of direct comparisons between preoperative and postoperative approaches.

Funding

National Key Research and Development Program of China, National Natural Science Foundation of China, Beijing Natural Science Foundation, Capital's Funds for Health Improvement and Research, National High Level Hospital Clinical Research Funding, and Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences.

Research in context.

Evidence before this study

We systematically searched PubMed, EMBASE, Cochrane Library, and Web of Science from inception to April 2025 for existing evidence on intravenous (IV) iron therapy, without language restrictions. Key terms included “intravenous iron OR IV iron OR iron supplementation” AND “surgery OR surgical patients” AND “postoperative OR preoperative OR intraoperative OR timing”. We identified multiple meta-analyses addressing the topic. The impact of IV iron on transfusion rates and postoperative haemoglobin levels differs across studies and may be linked to the timing of administration. Importantly, there are no data comparing the efficacy of IV iron between preoperative and postoperative supplementation.

Added value of this study

To our knowledge, this is the first network meta-analysis to compare preoperative versus postoperative IV iron supplementation. Our findings suggest that postoperative IV iron supplementation may be most effective in reducing transfusion rates, while preoperative supplementation may significantly improve haemoglobin recovery, particularly one month after surgery. Our findings also suggest that the benefits of IV iron supplementation depend on the timing.

Implications of all the available evidence

Our results may help guide the timing of IV iron administration based on the patient's specific needs. More randomised controlled trials are needed to directly compare outcomes between preoperative and postoperative IV iron supplementation. Furthermore, studies should investigate whether administering IV iron both before and after surgery results in improved outcomes for surgical patients.

Introduction

Anaemia can increase postoperative complications and prolong hospital stays in surgical patients.^1,2 Therefore, addressing anaemia is crucial for optimizing outcomes. Since anaemia frequently arises from iron deficiency,³ intravenous (IV) iron supplementation has emerged as a strategy in managing anaemia among surgical patients.⁴ It may even have a role in prehabilitation programs.⁵

Current evidence suggests that IV iron administration may increase postoperative haemoglobin levels, decrease transfusion volumes, shorten hospital length of stay, and lower the risk of nosocomial infections.6, 7, 8 Stratifying studies by the timing of iron supplementation (i.e., preoperative, postoperative, or perioperative) also yielded interesting outcomes. Two meta-analyses have shown that preoperative IV iron therapy can significantly reduce the rate of allogeneic blood transfusion.^9,10 The efficacy of postoperative iron supplementation in this aspect has also been demonstrated.^11,12 In addition, a 2017 study reviewed the literature on IV iron supplementation during the preoperative, perioperative, and postoperative periods; the authors concluded that the most substantial evidence for IV iron supported its use in the preoperative setting, while the decision to administer IV iron preoperatively or postoperatively remained individualised.¹³

Currently, there is a lack of randomised controlled trials (RCTs) or meta-analyses comparing preoperative and postoperative iron supplementation, making it difficult to determine the optimal timing. Therefore, this systematic review and network meta-analysis aims to compare preoperative versus postoperative iron supplementation in terms of efficacy, safety, and outcomes.

Methods

Search strategy and selection criteria

This systematic review and network meta-analysis was drafted following the Preferred Reporting Items for Systematic Reviews-Network Meta-Analyses (PRISMA-NMA) guideline and the Cochrane Handbook of Systematic Reviews of Interventions.^14,15 We prospectively registered our review protocol on the International Prospective Register of Systematic Reviews (PROSPERO, https://www.crd.york.ac.uk/PROSPERO/) on August 29, 2024 (CRD42024533265).

We searched PubMed, EMBASE, Cochrane Library, and Web of Science from inception to May 1, 2025, for randomised controlled trials (RCTs) investigating the administration of intravenous (IV) iron to surgical patients. The search strategy for each database is outlined in the Supplementary appendix (pp 1–2). To ensure that all relevant studies were identified, we manually checked references of similar systematic reviews and found no additional records.6, 7, 8, 9, 10, 11, 12, 13^,16, 17, 18, 19, 20, 21 Only publications in English were included.

Two reviewers (C Liu and R Fu) independently conducted the study selection. Only publications fulfilling the following Population, Intervention, Comparator, Outcome, and Study Design (PICOS) criteria were included: (1) population: patients undergoing surgery; (2) intervention: IV iron administered either 7–30 days before surgery or 0–30 days after surgery, with a total dose ranging from 300 mg to 3 g; (3) comparator: placebo, standard care, or oral iron; (4) outcome: articles that reported at least one of the primary or secondary outcomes (detailed below); (5) study design: RCT. Exclusion criteria included: (1) patients under 18 years old, pregnant women, critically ill patients, and patients who received prior transfusions; (2) unspecified timing for iron supplementation, administration both preoperatively and postoperatively, and any dose given outside the defined time frames of this study; (3) any intervention other than IV iron. In cases of dispute, a third reviewer (J Han) was consulted to reach a final consensus.

Data analysis

Two reviewers (C Liu and R Fu) independently used a predesigned data collection form to extract summary-level data from the included studies. Differences were resolved by the third reviewer (T Li). Extracted data included: (1) study characteristics, including first author, country, year of publication, and registration number; (2) study design, including sample size and outcomes; (3) patient characteristics, including age, sex, type of surgery, and presence of anaemia; (4) details of the intervention and control, including dosage and timing; (5) prespecified primary outcomes and secondary outcomes; (6) data regarding the risk of bias.

The primary outcomes are: (1) proportion of patients who received blood transfusion; (2) change between the baseline haemoglobin level and the haemoglobin level on postoperative day (POD) 7 and POD30. The secondary outcomes include iron metabolism markers (changes in ferritin and transferrin saturation from baseline to POD7), length of hospital stay, complications, drug-related adverse events, mortality, and quality of life.

The risk of bias for each study was assessed using the Cochrane Risk of Bias assessment tool which includes the following six domains: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and others.^15,22 Each study was labeled as having low, unclear, or high risk of bias.

We conducted a frequentist network meta-analysis using the random effects model to perform indirect comparisons between preoperative and postoperative IV iron supplementation since a direct comparison was not available in the selected studies. Dichotomous outcomes were presented as relative risks (RR) and continuous outcomes were presented as mean differences (MD). Both were reported with their corresponding 95% confidence intervals (95% CI). For each primary and secondary outcome (see Supplementary Table S1 for more details), forest plots and league tables were used to compare results among different treatments. Heterogeneity was assessed using Higgins I² for each pairwise comparison. Sensitivity analyses were conducted through study exclusion, alternative statistical models, and subgroup analyses to test the robustness of the results. We performed subgroup analyses based on surgery type, with dedicated evaluation of abdominal, orthopedic, and cardiac surgical populations. To rank the three treatments (i.e., preoperative therapy, postoperative therapy, and control), we derived ranking probabilities from Monte Carlo simulation with 10000 iterations and applied surface under the cumulative ranking curve (SUCRA) scores to demonstrate the probability of each being the most effective intervention.²³ All statistical analyses were carried out using the Network and mvmeta packages in Stata 14.2 (Stata Corporation, College Station, Texas, USA).

Role of the funding source

The funders of this study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. All authors had full access to the data in the study and the corresponding authors (C Lin and WW) had final responsibility for the decision to submit for publication.

Results

Study selection is illustrated in Fig. 1. A total of 3443 publications were initially identified from EMBASE, PubMed, Cochrane Library, and Web of Science. After removing 1139 duplicate records, we excluded an additional 2174 studies after title and abstract screening. Subsequently, 107 studies were removed after full text review due to the following reasons: no full text available (n = 25), ineligible population (n = 11), ineligible interventions or control (n = 31), inappropriate timing of administration (n = 33), incorrect study type (n = 3), duplicate results from the same RCT (n = 4). Ultimately, 22 articles met the inclusion criteria and were included in the meta-analysis.24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45

Fig. 1.

Study selection. PRISMA flow diagram detailing study selection process of the study. RCT, randomised controlled trial. PRISMA, Preferred Reporting Items for Systematic Reviews.

The included studies were published from 2001 to 2025 across 12 countries. Eight compared preoperative IV iron with the control group,24, 25, 26, 27, 28, 29, 30, 31 while 14 compared postoperative IV iron with non-iron supplementation.32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 No direct comparison between preoperative and postoperative administration was found. A total of 3026 patients were included, with sample sizes ranging from 26 to 487. The mean age of the participants was 65.8 years (range 46.2–74.7). 53.6% (1623/3026) patients included in all studies were women (range 9.7%–92.6%). Types of surgery included cardiac, abdominal, and orthopedic procedures. The IV iron formulations consisted of iron sucrose, ferric carboxymaltose, ferric derisomaltose, and iron polymaltose. Further study characteristics are presented in Table 1.

Table 1.

Characteristics of all included studies.

Author (Country)	Year	Registration	Surgery	Inclusion criteria	Sample size (IV iron, control)	Attrition	Female sex	Mean age	Intervention	Control	IV iron timing
Akshata (India)32	2023	–	Elective orthopedic and general surgery	≥18 years old	60 (30, 30)	0	36 (60%)	50.7	IS, 200 mg∗3 doses	Saline, 200 mL∗3 doses	Postoperative (POD1, 3, 7)
Assouline (Switzerland)33	2021	NCT02631980	Elective liver surgery	≥18 years old	49 (24, 25)	0	11 (22.4%)	65.5 ± 8.5	FCM, 15 mg/kg (max 1 g)	Saline, same volume	Postoperative (4 h after surgery)
Bisbe (Spain)34	2014	EudraCT 2010-023038-22; NCT01913808	TKA	Postoperative anaemia (Hgb 85–120 g/L)	122 (60, 62)	18 (14.8%)	97 (79.5%)	72.8 ± 9.7	FCM, calculated to correct TID	Oral ferrous sulfate, 100 mg	Postoperative (POD1)
Choi (Korea)35	2022	NCT03561480	TKA	Postoperative anaemia (Hgb < 100 g/L)	109 (54, 55)	0	97 (89.0%)	71.6 ± 5.9	FCM, 500 or 1000 mg (according to weight)	None	Postoperative (POD3)
Edwards (UK)24	2009	2005-003608-13UK	Colorectal surgery	≥18 years old	60 (34, 26)	2 (3.3%)	21 (35%)	68.3	IS, 300 mg∗2 doses	Saline, 250 mL∗2 doses	Preoperative (14 days before surgery, 24 h interval)
Fung (China)25	2022	NCT03565354	Colorectal surgery	Preoperative IDA (Hgb < 130 g/L; ferritin< 30 μg/L or 30–100 μg/L with TSAT < 20%)	40 (20, 20)	0	16 (40%)	69.1 ± 10.0	IIM, 20 mg/kg (max 1 g)	Standard care	Preoperative (3 weeks before surgery)
Houry (Lebanon)36	2023	NCT03759964	Cardiac surgery	Non-anemic (Hgb ≥ 120 g/L)	195 (97, 98)	0	50 (25.6%)	65.5 ± 11.5	FCM, 1 g	Saline, 100 mL	Postoperative (POD1)
Khalafallah (Australia)37	2016	ACTRN12614001261606	Elective major surgery	Postoperative IDA (Hgb 70–120 g/L; ferritin ≤ 100 μg/L or TSAT ≤ 20%)	201 (103, 98)	10 (5.0%)	115 (57.2%)	66.0 ± 12.6	FCM, 15 mg/kg (max 1 g)	Standard care	Postoperative (≥1 day after surgery)
Madi-Jebara (Lebanon)38	2004	–	Elective cardiac surgery	Postoperative anaemia (Hgb 70–100 g/L)	61 (30, 31)	0	6 (9.8%)	57.3 ± 9.4	IS, calculated to correct TID (200 mg/day)	Placebo	Postoperative (from POD1 until TID reached)
Mudge (Australia)39	2012	ACTRN12608000186358	Kidney transplantation	≥18 years old without iron overload (TSAT ≤ 50%, ferritin ≤ 800 μg/L)	102 (51, 51)	4 (3.9%)	28 (27.5%)	46.4 ± 12.7	Iron polymaltose, 500 mg	Oral ferrous sulfate, 210 mg iron daily	Postoperative (POD4)
O'Loughlin (Australia)40	2023	ACTRN12612000448842	Surgery for femoral neck or subtrochanteric fracture	>60 years old with preoperative anaemia (Hgb female < 120 g/L, male < 130 g/L)	143 (70, 73)	0	96 (67.1%)	85.5 ± 8.2	FCM, calculated to correct TID (max 1 g)	Saline, 50 mL	Postoperative (at skin closure)
Olijhoek (Netherlands)26	2001	–	Orthopedic surgery	Anemic (Hgb 100–130 g/L) but non-iron-deficient (TIBC ≥ 15%, ferritin ≥ 50 μg/L)	52 (25, 27)	5 (9.6%)	46 (88.5%)	66.4 ± 12.6	IS, 200 mg∗2 doses	Oral iron, 200 mg daily∗14 days	Preoperative (14 days and 7 days before surgery)
Shokri (Egypt)27	2022	NCT04061655	CABG	52–67 years old with anaemia (Hgb female < 120 g/L, male < 130 g/L)	80 (40, 40)	0	37 (46.3%)	59.2 ± 4.7	FCM, 1000 mg	Saline, 100 mL	Preoperative (7 days before surgery)
Talboom (Netherlands)28	2023	NCT02243735	Colorectal surgery	≥18 years old with IDA (Hgb female < 120 g/L, male < 130 g/L; TSAT<20%)	202 (96, 106)	3 (1.5%)	97 (48.0%)	71.0 ± 13.5	FCM, 1–2 g (according to Hgb and weight)	Ferrous fumarate, 200 mg tid	Preoperative (4 weeks before surgery)
Thin (Singapore)29	2021	NCT03295851	Elective abdominal surgery	≥21 years old with IDA (Hgb female < 120 g/L, male < 130 g/L; ferritin < 100 μg/L or ferritin 100–300 μg/L with TSAT < 20%)	26 (13, 13)	4 (15.4%)	18 (69.2%)	59.2 ± 18.3	FCM 15 mg/kg (max 1 g)	Ferrous fumarate, 200 mg bid	Preoperative (1–4 weeks before surgery)
Xu (China)41	2019	ChiCTR1800014776	Cardiac valvular surgery	20–70years old, preoperative non-anemic, postoperative IDA (Hgb female < 120 g/L, male < 130 g/L; ferritin 30–100 μg/L or TSAT < 20%)	150 (75, 75)	0	83 (55.3%)	54.4 ± 11.2	IS, calculated to correct TID (200 mg/dose)	Saline, same volume	Postoperative (from POD1, every other day until TID reached)
Yoo (Korea)42	2021	NCT03470649	TKA	≥19 years old	89 (44, 45)	11 (12.4%)	67 (75.3%)	70.5 ± 6.5	FDI, 1–2 g (according to Hgb and weight, max 20 mg/kg)	Saline, 100 mL	Postoperative (at skin closure)
Kim (Korea)43	2017	NCT01725789	Gastrectomy	≥20 years old with postoperative anaemia (Hgb 70–100 g/L)	454 (228, 226)	24 (5.3%)	249 (54.8%)	61.1 ± 13.1	FCM, 500 or 1000 mg (according to weight)	Saline, 100 or 200 mL (according to weight)	Postoperative (about POD 5–7)
Kim (Korea)44	2021	NCT03561480	TKA	Postoperative anaemia (Hgb < 100 g/L)	118 (58, 60)	2 (1.7%)	105 (89.0%)	69.8 ± 4.4	FCM, 500 or 1000 mg (according to weight)	Saline, 100 or 200 mL (according to weight)	Postoperative (about POD0)
Richards (UK)30	2020	ISRCTN67322816	Major open abdominal surgery	≥18 years old with preoperative anaemia (Hgb female < 120 g/L, male < 130 g/L)	487 (244, 243)	13 (2.7%)	267 (54.8%)	65.5 ± 13.9	FCM, 1000 mg	Saline, 100 mL	Preoperative (about 12–22 days before surgery)
Keeler (UK)31	2017	UK 2011-002185-21; NCT01701310	Colorectal surgery	Anaemia (Hgb female < 110 g/L, male < 120 g/L)	116 (55, 61)	6 (5.2%)	44 (37.9%)	74.3 ± 9.0	FCM, 1–2 g (according to Hgb and weight, max 1 g/dose)	Oral ferrous sulfate, 200 mg bid	Preoperative (about 2 weeks before, some with a second dose 7 days before surgery)
Kremke (Denmark)45	2025	EudraCT 2020-001389-12; NCT04608539	Cardiac surgery	Moderate postoperative anaemia (Hgb 80–110 g/L)	110 (57, 53)	6 (5.5%)	37 (33.6%)	70.5 ± 11.3	FDI, 20 mg/kg	Oral ferrous sulfate, 100 mg bid∗4 weeks	Postoperative (POD1)

Data are N (n₁,n₂), n (%), or mean ± SD. SD is not presented if not provided by the included study. Interventions are given in a single dose unless otherwise specified.

UK, United Kingdom. TKA, total knee arthroplasty. CABG, coronary artery bypass graft. IDA, iron deficiency anaemia. Hgb, haemoglobin. TSAT, transferrin saturation. TIBC, total iron binding capacity. IS, iron sucrose. FCM, ferric carboxymaltose. FDI, ferric derisomaltose. TID, total iron deficit, calculated using the Ganzoni formula⁴⁶ {TID[mg] = body weight[kg] × (target Hb[g/dL]-actual Hb[g/dL]) × 2.4 + iron stores[mg]}. POD, postoperative day. IV, intravenous. bid, bis in die (twice a day). tid, ter in die (three times a day).

Nineteen studies^24,25,28, 29, 30, 31^,33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45 with 2847 participants included transfusion rate as an outcome of interest. Compared to the control, postoperative IV iron supplementation effectively reduced the transfusion rate (RR 0.80, 95% CI 0.68–0.94; I² = 0.0%), while preoperative IV iron did not cause a significant reduction (RR 0.91, 95% CI 0.72–1.15; I² = 0.0%) (Fig. 2 and Supplementary Figure S2). An indirect comparison between preoperative versus postoperative IV iron showed no significant difference between the two (RR 0.88, 95% CI 0.66–1.17), consistent with the league table (Supplementary Figure S1). The ranking probability and SUCRA scores both suggested that the postoperative group had the lowest transfusion rate (SUCRA 90.5), followed by the preoperative group (48.4) and control group (11.1) (Fig. 2 and Supplementary Table S2). Subgroup analysis stratified by surgery type revealed that both preoperative and postoperative IV iron did not significantly reduce transfusion rates after major abdominal surgery. Rather, cardiac surgery accounted for the majority of observed reductions in transfusion rates in the postoperative studies (Supplementary Figure S3).

Fig. 2.

Transfusion rate. (A) Forest plot for the network meta-analysis of transfusion rate, comparing preoperative IV iron, postoperative IV iron, and control, with pooled data from 19 studies. (B) Ranking probability of transfusion rate. The probability of ranking the 1st, 2nd, and 3rd for each approach. RR, risk ratio. CI, confidence interval.

Six studies,^{24,28,32,33,41,42} comprising 586 patients, administered all interventions before POD 7 and measured changes in haemoglobin. We calculated the changes in haemoglobin level from baseline using mean differences and standard deviation. Neither preoperative (MD 4.64, 95% CI −4.49 to 13.78; I² = 49.6%) nor postoperative (MD −4.51, 95% CI −9.75 to 0.72; I² = 90.3%) IV iron supplementation had a significant impact on haemoglobin levels by POD7 compared to the control group (Fig. 3 and Supplementary Figure S4). Sensitivity analysis revealed that one study (Akshata, 2023³²) was a major source of heterogeneity (I² = 9.8% after its exclusion) for postoperative studies (Supplementary Figure S5). However, the direction and significance of the pooled effect remained consistent. The ranking probability and SUCRA scores suggested that preoperative IV iron supplementation had the most impact on haemoglobin levels (SUCRA 90.0), followed by control (55.5) and postoperative IV iron (4.4) (Fig. 3 and Supplementary Table S2).

Fig. 3.

Haemoglobin levels. (A) Forest plot for the network meta-analysis of changes in haemoglobin level at POD7, comparing preoperative IV iron, postoperative IV iron, and control, with pooled data from six studies. (B) Ranking probability of haemoglobin at POD7. The probability of ranking the 1st, 2nd, and 3rd for each approach. (C) Forest plot for the network meta-analysis of changes in haemoglobin level at POD30, with pooled data from seven studies. (D) Ranking probability of haemoglobin at POD30. The probability of ranking the 1st, 2nd, and 3rd for each approach. MD, mean difference. CI, confidence interval. Hgb, haemoglobin. POD, postoperative day.

Seven studies involving 905 participants reported the changes in haemoglobin level at POD30.^27,28,35, 36, 37, 38^,42 Both preoperative (MD 12.11, 95% CI 7.88–16.35; I² = 85.7%) and postoperative IV iron supplementation (MD 5.45, 95% CI 2.70–8.20; I² = 45.5%) were significantly more effective than the control in maintaining haemoglobin levels 30 days after surgery, with the preoperative group outperforming the postoperative group (MD 6.67, 95% CI 1.61–11.72) (Fig. 3 and Supplementary Figure S4). The ranking probability also suggested that preoperative IV iron supplementation (SUCRA 99.7) was the optimal option, followed by postoperative IV iron (50.3) and control (0.0) (Fig. 3 and Supplementary Table S2). Due to the limited number of studies in each category, formal subgroup analyses were not performed for both haemoglobin outcomes to avoid unreliable estimates (Supplementary Figure S5).

Several studies also reported metrics such as haemoglobin responders (defined as an increase of >20 g/L from baseline), time to anaemia resolution, and anaemia incidence. Among studies comparing postoperative and control groups, one study revealed a significantly higher proportion of haemoglobin responders in the intervention group two weeks after surgery (P < 0.001), although this difference gradually diminished over time and became nonsignificant at postoperative week eight (P = 0.95).³⁵ Conversely, another study found that the intervention group maintained a significant difference 12 weeks after surgery (P = 0.001).⁴³ Three studies reported a shorter time to anaemia resolution in the postoperative intervention group, although this was not statistically significant compared to the control group.^35,39,44 Additionally, decreased anaemia incidence was seen at POD30 after both preoperative (P < 0.001) and postoperative IV iron therapy (P = 0.036) compared to control.^27,42

Thirteen studies reported indicators of iron metabolism, such as serum iron, ferritin, TSAT, and hepcidin, among others.^24,28,32, 33, 34, 35, 36, 37, 38^,41, 42, 43, 44 We chose to further investigate ferritin and TSAT, the two most commonly reported markers, at POD7 for this meta-analysis. Eight studies including 788 participants reported ferritin levels at POD7.^{24,28,32,33,36,38,41,42} Postoperative IV iron supplementation was significantly more effective in increasing ferritin levels compared to control (MD 743.83, 95% CI 440.51–1047.15; I² = 95.9%) but not preoperative therapy (MD 448.30, 95% CI −153.84 to 1050.44) (Fig. 4 and Supplementary Figure S6). Sensitivity analyses showed robust heterogeneity when excluding any single study. Consistently, postoperative therapy ranked first among the three groups (SUCRA 96.2), followed by preoperative therapy (47.0) and control (6.8) (Supplementary Table S2 and Supplementary Figure S9).

Fig. 4.

Secondary outcomes. Forest plots for the network meta-analysis of (A) changes in ferritin level at POD7. (B) Changes in TSAT level at POD7. (C) Length of hospital stay. (D) Complications. (E) Drug-related adverse events. (F) Mortality. MD, mean difference. CI, confidence interval. RR, risk ratio. POD, postoperative day. TSAT, transferrin saturation.

TSAT at POD7 was reported in six studies, comprising 528 patients.^{24,28,32,33,41,42} The postoperative group showed a significant increase in TSAT compared to the control group at POD7 (MD 13.37, 95% CI 1.33–25.42; I² = 93.1%), whereas the preoperative group did not show any increase in TSAT (MD −4.75, 95% CI −21.31 to 11.82; I² = 0.0%) (Fig. 4 and Supplementary Figure S6). Excluding one study (Yoo 2021)⁴² substantially reduced heterogeneity (I² from 93.1% to 19.0%, Supplementary Figure S6). The SUCRA scores also proved postoperative iron therapy to be the optimal option in elevating TSAT (SUCRA 97.1), followed by the control (36.1) and preoperative group (16.9) (Supplementary Table S2 and Supplementary Figure S9).

The length of hospital stay was reported in 14 studies involving 1869 participants.^24,25,27, 28, 29, 30, 31^,33,34,37,40, 41, 42^,45 No significant difference was found in the three pairwise comparisons (Fig. 4 and Supplementary Figure S8). According to the SUCRA score, the two interventions ranked very closely and were slightly better than the control (Supplementary Table S2 and Supplementary Figure S9).

Ten studies involving 1539 patients reported the incidence of complications after surgery.^{25,27,28,30,35,37,39,40,42,44} In most studies, complications referred to wound or systemic infections. No significant difference was found in the three pairwise comparisons (Fig. 4 and Supplementary Figure S7). Ranking probability showed that postoperative IV iron supplementation was the optimal option in reducing the risk of surgical complications (SUCRA 81.9) compared to preoperative therapy (57.9) and control (10.2) (Supplementary Table S2 and Supplementary Figure S9).

14 studies, including 1988 patients, reported the incidence of drug-related adverse events.^26,28,29,31,34, 35, 36, 37, 38, 39^,41,43, 44, 45 The incidence was not significantly higher in the two intervention groups, and no significant difference was found between preoperative and postoperative IV iron therapies (Fig. 4 and Supplementary Figure S7). The rankings of the three groups were control (SUCRA 59.6), preoperative therapy (48.7), and postoperative therapy (41.7) (Supplementary Table S2 and Supplementary Figure S9). Among the 14 studies, seven, involving 846 participants, reported no injection-site infection in either group.^29,35, 36, 37, 38^,41,45 One study reported five injection-site infections in the IV iron group (n = 228, 2.2%) and zero in the control group (n = 226, 0%).⁴³

A total of nine studies with 1486 patients investigated mortality after surgery.26, 27, 28^{,30,31,36,38,40,41} There were no significant differences across all three groups on pairwise comparison (Fig. 4 and Supplementary Figure S8). The two interventions ranked very closely and were slightly better than the control (Supplementary Table S2 and Supplementary Figure S9).

A total of nine included studies^{25,29,30,34,35,37,}43, 44, 45 and two subsequent studies^47,48 documented outcomes related to quality of life, assessed by measures such as Quality of Recovery Scale-15 (QoR-15), EuroQol Five Dimensions Questionnaire (EQ-5D), and 36-Item Short Form Survey (SF36). Studies generally showed no differences between preoperative iron supplementation versus control.^25,29,30 Notably, however, subsequent studies reported that the preoperative resolution of anaemia was closely associated with long-term quality of life, 5-year overall survival, and 5-year disease-free survival.^31,47,48 Similarly, most studies comparing postoperative interventions versus control did not yield significant differences.^34,35,37,43, 44, 45 Only two studies demonstrated the advantage of iron therapy in the subscores of “usual activities”³⁴ and “role-physical”.³⁷ Furthermore, one study documented outcomes related to rehabilitation, such as the Barthel test and the 6-min walk test, but revealed no significant difference.³⁴

According to the risk of bias assessment (Supplementary Figures S10 and S11), six studies (6/22) were at low risk of bias across all domains.^{30,33,37,40,42,45} All included RCTs described random sequence generation, and more than 70% conducted allocation concealment and blinding of participants and personnel (17/22 and 16/22, respectively). While the majority of studies had a moderate risk of bias, only nine RCTs performed blinding of outcome assessment,^{27,28,30,33,35,37,40,42,45} indicating mixed quality of evidence across the included studies.

Discussion

This network meta-analysis of 222 studies and 3026 surgical patients compares preoperative and postoperative IV iron supplementation through indirect methods. We identified four main findings: 1) Postoperative iron supplementation was associated with reduced transfusion rates, increased haemoglobin levels at POD30, and increased ferritin and TSAT levels at POD7. 2) Preoperative iron supplementation led to even higher haemoglobin levels at POD30 compared to postoperative supplementation but was ineffective at reducing transfusion rates, improving haemoglobin levels, and improving iron metabolism markers at POD7. 3) IV iron supplementation did not significantly reduce hospital length of stay, complications, or mortality rates and did not improve quality of life. 4) IV iron supplementation did not pose a higher risk of drug-related adverse events.

To evaluate the efficacy of IV iron, we used transfusion rate and haemoglobin level as the primary outcomes of this study, as they are the two most commonly employed metrics. Regarding transfusion rate, some prior meta-analyses suggest that preoperative IV iron supplementation effectively lowers autologous blood transfusion rates,^9,10,49 but also with exceptions.¹⁹ Our study differed from these findings primarily due to differences in study design, especially regarding the definition of the “preoperative” period. We defined “preoperative” as extending beyond one week before surgery, whereas many other meta-analyses blurred the line between short-term and long-term preoperative periods. In contrast, our study findings regarding postoperative iron supplementation aligned with prior studies; specifically, postoperative IV iron was associated with a significant reduction in transfusion rates.^11,12

It is generally believed that IV iron can increase ferritin and haemoglobin levels, regardless of whether the individual is anemic or not.⁵⁰ For example, previous studies suggest that both preoperative and postoperative iron supplementation contribute to the postoperative recovery of haemoglobin levels, especially past four weeks after surgery.^9,11 Another study showed that compared to oral iron, postoperative IV iron supplementation has a significant advantage in increasing haemoglobin levels past two to four weeks after surgery.⁵¹ In our study, postoperative iron supplementation increased ferritin and TSAT levels on POD7 but did not raise haemoglobin levels. By POD30, postoperative IV iron notably enhanced haemoglobin levels, although preoperative supplementation showed superior improvement. Of note, IV iron often requires more than one week to exert its full effect on haemoglobin levels. Our findings also suggest that after surgery, the body may initially focus on replenishing iron stores before boosting haemoglobin levels with the supplemented iron.^28,36,38 The time required for the body to convert iron into haemoglobin also explains why preoperative IV iron was more effective than postoperative IV iron at increasing haemoglobin levels.

In summary, postoperative IV iron supplementation was better at reducing transfusion rates, while preoperative supplementation led to improved haemoglobin levels. One possible explanation is that haemoglobin recovery is a delayed effect of preoperative iron supplementation, so its benefits may not be seen in the early postoperative period when transfusions are often needed. In contrast, though less effective for long-term haemoglobin recovery, postoperative iron supplementation may buffer the transient sharp decline of haemoglobin level immediately after surgery.

Regarding safety, two meta-analyses reported an increased risk of infection with IV iron supplementation compared to oral iron or no iron.^52,53 Meanwhile, another study involving 103 trials reported that IV iron therapy is not associated with an increased risk of severe adverse events or infections and can even reduce gastrointestinal adverse events.⁵⁴ Similar with other studies on surgical RCTs,^6,9,17 our meta-analysis found no statistically significant increase in the risk of drug-related side effects with both preoperative and postoperative IV iron. The existing evidence from our included studies also indicates a low incidence of injection-site infection. One potential explanation is that surgical RCTs may overlook injection-site infections and prioritise surgical site infections and systemic infections. This reporting bias may lead to underestimation of injection-site infections in the literature. In terms of our results so far, there is no evidence that IV iron is unsafe in surgical patients, regardless of the timing of administration.

Regarding postoperative recovery, the majority of studies,^19,49,53 except for one,¹⁰ did not report any impact of IV iron supplementation on mortality or hospital length of stay. In non-anemic patients, IV iron was shown to enhance physical function and alleviate fatigue without an improvement in overall quality of life.⁵⁰ In surgical patients, postoperative iron supplementation, whether IV or oral, was not found to impact quality of life or functioning.⁵¹ These findings, which largely align with ours, suggest that IV iron supplementation offers limited benefits for postoperative recovery and rehabilitation.

Of note, most previous studies define “preoperative” as any time before surgery and include many RCTs that administer IV iron only one to three days in advance. However, evidence shows that IV iron requires approximately seven days to increase haemoglobin by 1 g/dL and about a month for a 2 g/dL increase.⁷ Thus, we defined the preoperative group as receiving iron 7–30 days before surgery to ensure that patients are all adequately prepared with iron before the operation,⁵⁵ enabling us to make comparison between preoperative preparation (prehabilitation) and postoperative supplementation. This distinguishes our study from most other meta-analyses in this field.

One limitation of this study arises from differences in study design of the included RCTs, particularly in patient selection, dosing regimen, and outcome assessment. For example, studies regarding postoperative IV iron typically included patients with postoperative anaemia, while preoperative studies did not, leading to selection bias. The limited number of studies, especially preoperative trials, likely contributed to the observed heterogeneity. Additionally, insufficient data prevented comprehensive subgroup analyses. While subgroup analyses by type of surgery were attempted, the conclusions about transfusion rate require cautious interpretation. Other outcomes or subgroups (e.g., anaemia status and sex difference) lacked adequate data for analysis, highlighting the need for larger, standardized studies. Furthermore, the inclusion of both oral iron and placebo in the control group likely introduces bias to the study. However, it is worth noting that prior studies have demonstrated similar effects between the two.⁶ Ferritin as a secondary outcome also has limitations because it may transiently elevate after IV iron administration and during inflammation, independent of true iron status. Finally, the lack of studies directly comparing preoperative versus postoperative IV iron is a notable limitation of this study. Thus, readers should exercise caution in interpreting the relative effectiveness until direct comparisons are available.

Our findings have important implications for clinical practice in determining the ideal timing for IV iron supplementation in surgical patients. The Perioperative Patient Blood Management Guidelines recommend addressing anaemia and iron deficiency as early as possible but within a month before surgery.⁵⁶ Our findings support the prehabilitative use of IV iron to optimise postoperative haemoglobin levels and prevent severe anaemia. In addition, limiting blood transfusions can help decrease morbidity and mortality.56, 57, 58 Thus, we recommend the early use of IV iron after surgery to reduce transfusion needs, particularly in cases of postoperative anaemia. Finally, while IV iron is generally safe, potential side effects such as hypophosphatemia underscore the need for close monitoring during its administration.⁵⁹

IV iron administration shows promise in improving postoperative outcomes, but comparative studies on the optimal timing for administration are lacking. This indirect-comparison network meta-analysis suggests that postoperative IV iron reduces transfusion rates, while preoperative supplementation leads to better haemoglobin recovery, especially after one month. More RCTs are needed, especially those directly comparing the two therapies. Based on these findings, we propose that administering IV iron both before and after surgery may yield better clinical outcomes than either therapy in isolation. To validate this, future RCTs should compare this combined approach against placebo.

Contributors

C Liu, C Lin, and WW conceptualised the study and designed the methodology. C Liu, JH, RF, and TL were involved in formal analysis, and KM curated the data. C Liu wrote the original draft of the manuscript. C Lin, GM, and JW contributed to its review and editing. C Lin and WW supervised the study and acquired funding. C Liu, C Lin, and WW had access to and verified the underlying study data. All authors read and consented to the published version of the manuscript. All accept the responsibility to submit the manuscript for publication.

Data sharing statement

The data generated in this study are available from the corresponding authors upon reasonable request, for research purposes only. The data will be available starting from the date of publication.

Declaration of interests

The authors declare no competing interests.