2.1 Eligibility criteria

Studies were considered eligible for inclusion if they met the following criteria: patients ≥ 18 years of age who were undergoing a primary total hip (THA) or knee (TKA) arthroplasty; patients received a perioperative nutritional intervention designed by a dietitian; written in the English language; and study design was a randomized control trial (RCT), cohort, or case control.

The exclusion criteria included: systematic reviews, conference abstracts, case reports and case series study designs; patients undergoing revision THA or TKA, and exclusive use of supplements or drinks without perioperative nutritional intervention from a dietitian.

2.2 Search strategy and study selection

A systematic review of the literature was conducted in June 2022 using online databases “Ovid Medline”, “Embase”, and “Cochrane Library.” The search reviewed all fields of the available peer-reviewed literature published in the English language. Articles were identified using an electronic search of the following keyword terms and their respective combinations: “hip replacement or hip arthroplasty” or “knee replacement or knee arthroplasty” and “nutritional support” or “nutritional supplementation” or “enteral nutrition” or “malnutrition” or “immunonutrition” or “nutritional status” or “supplementation of iron or albumin or carbohydrate or amino acids” or “carbohydrate loading” or “dietary supplements” or “nutritional status” or “diet therapy.” Articles were reviewed for additional citations and abstracts were considered. A professional librarian conducted the search and performed an initial screening for duplicates. All articles were uploaded to Raayan (https://www.rayyan.ai/) for screening. Titles and abstracts that did not include hip and/or knee replacement and dietary counseling or dietitian-led nutritional intervention were excluded. We then screened full-text articles to determine whether inclusion criteria were met. Any disagreements between the reviewers were resolved through a third author until consensus was reached.

2.3 Assessment of methodologic quality and risk of Bias

Two reviewers independently performed methodological quality assessments of the selected articles based on the Methodological Index for Non-Randomized Studies (MINORS) and the Consolidated Standards of Reporting Trials (CONSORT). MINORS and CONSORT are instrument tools to assess the methodological quality of non-randomized studies and randomized studies, respectively. MINORS was used to evaluate the sole non-randomized study by Schroer et al., while the remaining studies were assessed using CONSORT. The MINORS average score for Schroer et al. was 20.5 ± 1.5 out of 24.¹⁹ The CONSORT average score for the three RCTs was 14.6 ± 1.65 out of 25.^20,21 Inter-rater reliability was “strong”, with a measured kappa coefficient of 0.83.²²

2.4 Data extraction

Two blinded reviewers independently screened titles and abstracts of articles that arose from the search. The reviewers then independently conducted the data extractions from each study using a pre-defined extraction sheet. Data on study design, number of patients, mean age (years), nutritional intervention, and LOS were collected from each study. In addition, C-reactive protein (CRP) levels, albumin transfusion rates, serum electrolyte levels, total time out of bed postoperatively, and hospitalization charges were collected if included in the study.

3.1 Study selection and characteristics

A summary of the four included study characteristics is shown in Table 1. The four studies were a randomized pilot study, two RCTs, and a prospective cohort study.^2,23–25 The four articles represented a total of 5006 individuals undergoing TJA, ranging from 32 to 4733 participants. 86 (1.7 %) were male, 178 (3.5 %) were female, and 4742 (94.8 %) gender was not given. The mean age reported was 60 years old (range 26–85). There were 111 THAs, 162 TKAs, and 4733 unspecified whether they received THA or TKA. Primary outcomes varied among studies, with LOS being the common outcome assessed in all four studies. Two investigations included malnourished patients^2,23 and the remaining two included only well-nourished patients^24,25

3.2 Length of stay

All four studies investigated the effect of perioperative nutritional intervention for primary TJA on patient LOS (Table 2). All four studies utilized protein supplementation^2,23–25; they each demonstrated a statistically significant difference in LOS, favoring the intervention group (outlined in Table 2). As each study employed different interventions and while a meta-analysis does not apply, the weighted average in reduction of LOS was 0.75 days between the intervention and control groups (3.28 days vs. 4.03 days).

3.3 Other outcomes

A summary of the secondary outcomes assessed amongst the four studies^2,23–25 are listed in Table 2. Lower postoperative CRP levels, albumin infusion rates, rates of hypocalcemia and hypokalemia, wound drainage, costs associated with primary hospitalization and hospital readmissions, and increased total time out of bed postoperatively were found to be significantly improved in TJA patients who underwent dietitian-led interventions compared to patients with conventional care.

4.1 Length of stay

LOS is a major driver of cost associated with TJA and is outcome measure common to all four studies. Alito et al. found the combined effect of preoperative dietitian assessment and a protein immune supplement with an oral carbohydrate solution five days prior to surgery in THA patients to have a significant reduction in median LOS (3 days, range 2–5 days vs 6 days, range 3–8 days, p < 0.01).²³ Cao et al. implemented a multimodal nutritional management regimen in which TKA patients received nutritional powder the day before surgery, early feeding starting two to 4 h after surgery, and had dietitian-led education and provision of carbohydrate, protein, and nutritional powder. They found a significantly decreased hospital LOS in this group (3.9 days ± 1.4 vs 5.6 days ± 3.3; P < 0.001) compared to controls.²⁴ Petersen et al. illustrated that THA patients who receive multimodal optimization, including calculation of daily fluid and energy intake, and a post-operative protein rich drink, had a moderate reduction in LOS (7.0 days, range 1–9 vs 8.0 days, range 1–10; p = 0.019) compared to controls without multimodal optimization. Lastly, Schroer et al. examined malnourished and nourished patient populations undergoing TJA across a network of five hospitals: four control and one study hospital during a 3-year period. Their dietary intervention was implemented during the last year of the study period only at the study hospital and consisted of a high-protein, high anti-inflammatory diet with a preoperative education class. Following attendance of the class, patients were instructed to begin this diet immediately and to follow it for at least one month after surgery. They showed a significant reduction in LOS (3.2 ± 1.4 days vs 3.9 ± 2.8 days, p = 0.04) in malnourished patients at the study hospital compared to malnourished patients at the control hospitals during the same period. Contrary to studies solely using nutritional supplementation alone, our results consistently revealed a reduction in LOS across interventions that incorporated a dietitian-led nutritional intervention. Notably, studies analyzing the impact of a carbohydrate beverage without a dietitian-led intervention observed no difference in LOS between their intervention and control groups in TJA patients.^15,16 This reduction in LOS through dietitian-led nutritional interventions suggests that improved nutritional status plays a role in mitigating postoperative complications linked with malnutrition, including hypoalbuminemia, electrolyte disorders, and PJI. However, it is worth noting that as TJA has migrated largely to same-day and 23-h discharge, LOS may no longer be as meaningful a metric to report.

4.2 C-reactive protein

CRP increases after surgery due to the acute phase reaction to trauma and can be used to monitor the inflammatory response following TJA.²⁸ Only Alito et al. studied postoperative CRP levels following TJA, particularly using immunonutrition. The acceleration of postoperative recovery (ACERTO) protocol group exhibited significantly decreased CRP values on the second postoperative day (66.5 mg/L ± 16.4 vs 80.6 mg/L ± 10.9, p < 0.01) compared to the control group. The authors suggested the reduction in CRP values of the ACERTO group were attributable to a shorter pre-operative fasting period and the use of the immune enhancing supplements, which include arginine, omega-3 fatty acids, and nucleotides, could modulate the inflammatory response as described in recent studies.^29,30 Prior research suggests supplements with immunomodulating properties decrease the incidence of infectious complications, improve the postoperative immunological response, and speed up recovery from immunodepression following surgical trauma.^31,32 Preliminary evidence suggests that dietitian-led assessment and education in the perioperative setting of TJA significantly reduces postoperative CRP levels. However, determining the optimal intervention requires further investigation. This analysis reinforces the potential of immunonutrtion, guided by dietitian expertise, in effectively managing postoperative inflammation.

4.3 Albumin and wound secretion

Given that malnutrition can lead to wound infection and delayed healing, Cao et al. selected serum albumin level as a biochemical marker for malnutrition.²⁴ Cao et al. found significant benefits with their nutrition regimen with regard to albumin levels and wound drainage.²⁴ Their intervention group displayed significantly lower rates of albumin infusion (25.9 % vs 45.2 %; P = 0.021), significantly less albumin required per infusion (6.5 ± 15.2 vs 15.2 ± 23.3 g; P = 0.006), and significantly lower wound drainage (9 vs 21 patients; p = 0.015) as compared to the controls. They also had a shorter preoperative fasting and earlier postoperative diet (2–4 h) compared to the controls. A better functional recovery was achieved due to a lower incidence of malnutrition and delayed wound healing as reflected through higher serum albumin levels following TJA. Importantly, the impact extends beyond the realm of TJA, as hypoalbuminemia has been associated in adverse outcomes across diverse surgical domains, such as cardiac surgery and surgical oncology.^33,34 One study identified hypoalbuminemia as a risk factor for delayed wound healing in patients undergoing abdominoperineal resection,³⁴ supporting further exploration into the potential benefits of correcting hypoalbuminemia to alleviate wound complications and optimizing surgical outcomes.

4.4 Electrolytes

Dietary intervention has been documented to reduce the incidence of electrolyte disorders following TJA, a critical consideration given the advanced age of patients to postoperative electrolyte imbalances in this demographic. The repercussions of electrolyte disorders on orthopaedic surgeries outcomes are well documented, including an increased risk of LOS, in-hospital myocardial infarction, surgical site infections, delirium, and readmission rates.^35–39 Notably, the MNM protocol employed by Cao et al. identified significantly lower rates of hypokalemia (17.3 % vs 33.3 %; p = 0.019) and hypocalcemia (60.5 % vs 76.5 %; p = 0.028), illustrating the potential of targeted dietary interventions in mitigating electrolyte imbalances. An additional study investigated the efficacy of preoperative oral electrolyte-carbohydrate nutrient supplement on electrolytes (sodium, calcium, and potassium) in primary elective TKA patients.⁴⁰ The rate of hyponatremia was reduced in the intervention (6.4 %) vs control group (21.3 %), yet no significant differences were found in regard to hypokalemia or hypocalcemia. While the specific correction of electrolyte disorders lacks a consensus and was not measured in the other studies included in this review, the collective evidence from these studies implies an impact of dietary interventions in averting postoperative electrolyte disorders after surgery. Thus, the administration of an electrolyte drink preoperatively, coupled with the consideration of a carbohydrate load in non-diabetic patients, stands as a strategy to proactively address this critical aspect of surgical care.

4.5 Mobilization and recovery

Petersen et al. tested the impact of a multimodal rehabilitation program on mobilization and nutrition following TJA.²⁵ By using aggressive postoperative mobilization plans and early fluid/diet re-introduction, they found a significantly increased average total time out of bed (37.4 h ± 10.4 vs 25.5 h ± 14.4; p < 0.001), increased average protein intake in the first 4 days postoperatively (1.25 g/kg ± 0.35 vs 0.74 g/kg ± 0.25; p < 0.0001), and improved mobilization efficiency (p < 0.001) compared to controls. Compared to conventional care, optimization of nutritional intervention through multimodal rehabilitation can lead to increases in mobilization because postoperative walking and nutritional goals were established preoperatively and taught to the patient.

4.6 Cost of care

Schroer et al. showed lower charges associated with primary hospitalization, hospital readmission and 90-day period (p < 0.001) in malnourished patients with nutritional intervention at the study hospital compared to malnourished patients without intervention at the control hospital. However, the hospital readmission rate within 90 days of surgery at the study hospital did not vary over the 3 year study period for either the nourished cohort (p = 0.06) or the malnourished cohort (p = 0.50) despite the introduction of a high-protein and anti-inflammatory diet during the last year.² The decreased cost of care may be explained by lower utilization of hospital services and the decreased LOS in those TJA patients enrolled in the dietary intervention and educational nutritional class.

This review highlights that a dietitian-led interventional program incorporating education with supplements, rather than supplements alone, results in more consistent benefits following TJA. Although our review provides valuable insight on the relationship of dietitian-led interventions on outcomes and complications, the studies included have limitations. First, only Schroer et al. study separates populations for comparison: malnourished and nourished patients (serum albumin <3.4 and >3.4 g/l respectively). The other studies had heterogenous patients identified as malnourished between their control and intervention groups, excluded individuals with a preoperative albumin <3.5 g/l, or did not report their baseline nutritional status. Yet, data supports that dietitian-led interventions improves TJA outcomes, primarily through decreasing LOS, regardless of the patient's nutritional status. Second, Alito et al. and Schroer et al. mention a follow-up period for their study's cohort while the remaining articles omitted this or did not have a follow-up period after discharge. Third, there is no consensus to how long patients need to optimize their nutrition prior to surgery.² Initiation of perioperative nutrition across studies in this review ranged from one day prior to surgery, five days prior to surgery, or preoperatively and continuation for at least one month after surgery. Nevertheless, research studying nutritional educational programs for patients has shown involving patients in their own nutritional care is an effective method to raise intake of energy and protein.⁴¹ Overall, the data suggests perioperative nutritional intervention has promising effects on patient outcomes and identifying these limitations may help direct future studies.