Impact of anaemia at discharge following colorectal cancer surgery

Objectives Preoperative anaemia is common in patients with colorectal cancer and increasingly optimised prior to surgery. Comparably little attention is given to the prevalence and consequences of postoperative anaemia. We aimed to investigate the frequency and short- or long-term impact of anaemia at discharge following colorectal cancer resection. Methods A dedicated, prospectively populated database of elective laparoscopic colorectal cancer procedures undertaken with curative intent within a fully implemented ERAS protocol was utilised. The primary endpoint was anaemia at time of discharge (haemoglobin (Hb) < 120 g/L for women and < 135 g/L for men). Patient demographics, tumour characteristics, operative details and postoperative outcomes were captured. Median follow-up was 61 months with overall survival calculated with the Kaplan-Meier log rank method and Cox proportional hazard regression based on anaemia at time of hospital discharge. Results A total of 532 patients with median 61-month follow-up were included. 46.4% were anaemic preoperatively (cohort mean Hb 129.4 g/L ± 18.7). Median surgical blood loss was 100 mL (IQR 0–200 mL). Upon discharge, most patients were anaemic (76.6%, Hb 116.3 g/L ± 14, mean 19 g/L ± 11 below lower limit of normal, p < 0.001). 16.7% experienced postoperative complications which were associated with lower discharge Hb (112 g/L ± 12 vs. 117 g/L ± 14, p = 0.001). Patients discharged anaemic had longer hospital stays (7 [5–11] vs. 6 [5–8], p = 0.037). Anaemia at discharge was independently associated with reduced overall survival (82% vs. 70%, p = 0.018; HR 1.6 (95% CI 1.04–2.5), p = 0.034). Conclusion Anaemia at time of discharge following elective laparoscopic colorectal cancer surgery and ERAS care is common with associated negative impacts upon short-term clinical outcomes and long-term overall survival. Electronic supplementary material The online version of this article (10.1007/s00384-020-03611-0) contains supplementary material, which is available to authorized users.

The prognostic value of preoperative anaemia in various CRC subgroups has been studied with reported associations between anaemia and adverse outcomes including decreased survival [5][6][7][8][9]. These findings underpin a number of clinical guidelines stating preoperative anaemia should be identified and corrected [10,11].
Comparatively little attention is given to the prevalence and consequences of postoperative anaemia with no available reports on its impact. Postoperative anaemia may influence functional recovery and overall patient experience including after discharge. A recent international consensus statement on anaemia management after major surgical procedures recommended correction prior to discharge [12] but there is no specific data following colorectal cancer surgery. We therefore aimed to investigate the frequency of postoperative anaemia and any associated short-and long-term implications following colorectal cancer resection.

Methodology
We performed an observational review of a dedicated prospectively compiled colorectal cancer patient database managed by a specialist information analytical team. Inclusion criteria were patients with biopsy proven colorectal cancer undergoing elective laparoscopic resection with curative intent between 2002 and 2015. Patient demography (age, sex, body mass index (BMI), American Society of Anaesthesiologists (ASA) grade) and histopathologically defined tumour staging data was reviewed. The primary endpoint was postoperative anaemia at time of discharge defined as haemoglobin (Hb) < 120 g/L for women and < 135 g/L for men using the last laboratory measure prior to hospital discharge. All cases began with a laparoscopic approach and were cared for within an established ERAS programme [13][14][15].
Secondary endpoints included preoperative anaemia (closest Hb measurement prior to surgery), perioperative time (from skin incision to completion of skin closure), total blood loss (volume of blood collected in suction systems and weighed swabs) and conversion to open surgery (inability to complete the dissection laparoscopically including vascular ligation(s) and/or requiring an incision larger than that required to remove the specimen [16,17]). Postoperative data included length of hospital stay (LoS, number of nights in hospital until discharge to home or care facility as appropriate with the day of surgery designated day zero), major postoperative complications (Clavien-Dindo grade III-V [18]) and readmission (unplanned hospital attendance within 30 days of discharge). All patients entered into a clinical, radiological and endoscopic surveillance follow-up programme for 5 years. Follow-up was defined as time of surgery to last clinical contact or death. This manuscript has been designed in accordance with the STROBE guidelines [19]. Initial creation and anonymised review of the database were approved by the local research ethics and data governance boards.
All data was explored for normality and displayed as mean ± standard deviations unless specified where medians with interquartile range are shown. Mean Hb below lower limit of the sex-specific normal range (LLN) was calculated for anaemic patient groups. t test, Mann-Whitney U and Kruskal-Wallis tests were used to compare medians from normal and non-normally distributed populations respectively. For categorical data, cross-tabulation and chi-squared testing assessed the difference between groups. Fisher's exact test was used when appropriate. The Kaplan-Meier log rank method was used to compare overall survival between groups. Cox proportional hazard regression assessed the effect of anaemia at discharge on survival probability simultaneously controlling for clinically relevant factors (age, sex, tumour location, stage, preoperative chemotherapy and postoperative complications). Diagnostics of independent variables confirmed that collinearity was not problematic (variance inflation factor < 2). Results are reported as hazard ratios (HR) with 95% confidence intervals. p < 0.05 was considered significant. The data was analysed using SPSS (v26, IBM, USA) and STATA (v8, StataCorp LLC, USA).

Discussion
Preoperative optimisation of anaemia has become well embedded within perioperative protocols but literature on postoperative anaemia and its sequelae remains scarce. Although an international consensus statement on postoperative anaemia management is available [12], the compliance with this guidance and the extent of the problem are unknown. Our study demonstrated the majority of our colorectal cancer patients were discharged anaemic and this impacted short-and long-term outcomes.
With no prior evidence for comparison, we report a potential association between anaemia and lower long-term overall survival. Although the exact mechanism(s) behind this link is unknown, this finding is comparable with CRC reports on preoperative anaemia and overall survival [8]. There is likely to be a wide array of interacting and contributing factors behind these observations including tumour cellular hypoxia and the ability to offer timely adjuvant chemotherapy and the subsequent response [4,20]. This concerning survival data is of importance given the very high rate of anaemia at discharge consistent with the limited available literature on other major procedures [21]. The precise causes of high postoperative anaemia rates among CRC patients are unclear. A significant proportion had preoperative anaemia which was not corrected prior to surgery and persisted through to discharge. This is particularly likely in the early phase of our cohort prior to our adoption of the current British Society of Haematology guidelines [10] although the rate of postoperative anaemia was higher than preoperative anaemia.
Intraoperative blood loss is another contributing factor to anaemia after surgery, but this was not a major cause in our patients and consistent with contemporary practice data including the lower volume blood loss associated with laparoscopic surgery [22][23][24]. Conversion to open surgery resulted Hospitalisation and postoperative complications are known risk factors for postoperative anaemia due to inadequate nutritional intake, malabsorption, frequent blood sampling and the systemic response to surgical stress [25][26][27]. This cohort was enrolled within an established ERAS programme. We have previously demonstrated that a significant proportion of patients deviate in more than one element of the postoperative ERAS pathway even in the absence of major complications resulting in longer LoS [28]. It is not clear if postoperative anaemia was a cause or a result from the delayed discharge and prolonged hospitalisation, and further studies are required to specifically address this issue as well as correcting postoperative anaemia and measuring the associated implications.
Traditional ERAS discharge criteria have strong emphasis on functional recovery metrics such as independent mobilisation, resuming oral intake and pain control with adequate analgesia. These discharge criteria could be modified to incorporate checking and correcting anaemia before discharge as this enhances functional recovery and the overall patient experience [29].
Although our pragmatic study is the first to explore anaemia at discharge and its outcomes, it has a number of limitations. Our findings should be interpreted with caution as these observations do not imply causation. Our study generates an interesting hypothesis worthy of further research to study the mechanism of this association and the external validity of these findings. Nevertheless, these findings suggest optimal patient care may involve the correction of anaemia before discharge.
We are unable to control for confounding factors such as comorbidities and the likely differing anaemia management and interventions that could influence our results and were likely to differ across the long study timeframe. The sample size limited meaningful sub-group analysis for relevant areas such as WHO criteria for severity of anaemia [30] and identifying a threshold figure to trigger anaemia correction. Both cancer specific and disease-free survival data is of interest but this was not routinely captured in these patients. Interestingly admission and discharge Hb levels were not associated with tumour stages in our cohort. We included only laparoscopic cases which had strict selection criteria particularly in the early phase of the study timeframe which may explain the relatively low number of advanced tumours in this cohort. Additionally, full data on the nature and severity of complications was incomplete particularly for the earliest patients and was not obtainable. Finally, our study design prevented capture of

Conclusion
Anaemia at time of discharge following elective laparoscopic colorectal cancer surgery is commonplace with associations upon short-term clinical outcomes and long-term overall survival. Correction of anaemia before leaving hospital may represent beneficial discharge criteria for future ERAS protocols.

Compliance with ethical standards
Conflict of interest NKF serves as chair of the UK ERAS Society. All authors confirm they hold no financial declarations or conflicts of interest for this paper.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.