Skip to main content
NCBI home page
Journal of Orthopaedics logoLink to Journal of Orthopaedics
. 2024 Jun 27;58:52–57. doi: 10.1016/j.jor.2024.06.041

The dynamics of perioperative serum albumin before and after total knee arthroplasty☆

Yoshinori Ishii a,⁎, Hideo Noguchi a, Junko Sato a, Ikuko Takahashi a, Hana Ishii b, Ryo Ishii c, Kai Ishii d, Shin-ichi Toyabe e
PMCID: PMC11266862  PMID: 39055285

Abstract

Background

Previous studies have evaluated preoperative serum albumin (SA) for predicting postoperative complications of total knee arthroplasty (TKA). This study aimed to investigate the dynamics of perioperative SA and changes in SA (ΔSA) and identify any influential patient or surgical factors.

Methods

In total, 381 patients (483 knees) undergoing primary TKA were recruited. SA values preoperatively (SA0), 1 week postoperatively (SA1W), and 4 weeks postoperatively (SA4W) were investigated. SA values were converted to a percentage of SA0 and differences between timepoints were calculated and expressed as follows: ΔSA1W−0, ΔSA4W–1W, and ΔSA4W−0. Patient and surgical factors previously identified or with the potential to influence SA were evaluated.

Results

The median values of SA0, SA1W, and SA4W were 4.4, 3.8, and 4.2 g/dL, respectively; SA0 was significantly different between groups (p < 0.001). The incidence of low SA0 (<3.5 g/dL) was less than 1 %. Median ΔSA values were −13.7 %, 9.6 %, and −4.5 % for ΔSA1W−0, ΔSA4W–1W and ΔSA4W−0, respectively; ΔSA was significantly different between groups (p < 0.001). SA4W recovered to 95.5 % of SA0 with less than 2 % of patients having low SA4W (<3.5 g/dL). Multiple regression analyses showed SA concentration at each timepoint was significantly associated with the other SA timepoint values; age was significantly associated with SA4W and SA1W (all p < 0.001).

Conclusions

We identified SA0 and age as significant factors affecting SA dynamics in the perioperative period. Low SA (<3.5 g/dL) was uncommon both preoperatively and at 4 weeks postoperatively; therefore, conventional cutoff values and preventive measures for low SA may need reconsideration.

Keywords: Preoperative serum albumin, Perioperative serum albumin, Total knee arthroplasty, Postoperative complications, Cutoff values

1. Introduction

Perioperative nutritional management has primarily been studied in the field of abdominal surgery; however, it is now a growing topic in the field of total joint arthroplasty (TJA). Of the possible biomarkers of preoperative nutritional status relevant to TJA, serum albumin (SA) appears to be the most informative.1, 2, 3 SA is the most effective screening tool for predicting increased postoperative complications of total knee arthroplasty (TKA).1,4,5 The main complications associated with low SA are increased postoperative wound infection rates,1,2,4,6, 7, 8, 9, 10 increased emergency room visits and readmission rates,5,6,11,12 and longer hospital stays.12

As a result of the concerns about increased perioperative complications and mortality associated with low SA,2,4, 5, 6,8,10 preoperative screening for hypoalbuminemia has been advocated.12, 13, 14 Torchia et al.15 concluded that universal, preoperative malnutrition screening and intervention among TKA candidates is a cost-effective way to ameliorate postoperative complications. However, most previous studies only evaluated preoperative SA values; no detailed examination of perioperative SA dynamics or the extent of SA changes (ΔSA) has been conducted to date.

Therefore, the purpose of this study was to investigate perioperative SA and ΔSA in TKA patients and identify any patient or surgical factors that may influence them. By characterizing perioperative SA trends and ΔSA, we aimed to identify patients who should receive preoperative intervention to reduce the risk of complications caused by low SA.

2. Materials and methods

This retrospective study was conducted using data collected at our institute between January 2001 and June 2023. Informed consent was obtained from all patients after discussion of the study, which included a description of the protocol. The institutional review board approved the study before its commencement. In total, data from 381 consecutive patients (483 out of 494 knees (98 %)) undergoing primary TKA, for whom all the necessary information for this study existed, was included from medical records. The preoperative diagnosis requiring TKA was osteoarthritis. SA values preoperatively (SA0), 1 week postoperatively (SA1W), and 4 weeks postoperatively (SA4W) were investigated. The differences in SA levels 1 week postoperatively and preoperatively, 4 weeks postoperatively and 1 week postoperatively, and 4 weeks postoperatively and preoperatively were defined as ΔSA1W−0, ΔSA4W–1W, and ΔSA4W−0, respectively, and their associations were analyzed. In addition, we examined the following factors that may influence SA: sex,4,5,8,14,16 age,4,5,8,14 body mass index (BMI),8,17 preoperative complications (hypertension, hyperlipidemia, and diabetes),4,8,10,14,16,18 and smoking history16 as patient factors, and clinical score (Hospital for Special Surgery score19), operative time,16 tourniquet time, and total blood loss10,16 as operative factors.

Patients who had undergone revision arthroplasties or previous tibial osteotomies, patients with rheumatoid arthritis, and patients with inadequate information were excluded. The clinical characteristics of the patients are summarized in Table 1. All surgeries were performed by a single surgeon using a standardized technique with the standard medial parapatellar approach, including the necessary soft tissue release for proper gap balancing with mechanical alignment principles under tourniquet control, which has been described in detail previously.20 All patients were allowed to start drinking water and eat after clear mentation was confirmed on the day of the surgery.

Table 1.

Patient characteristics.

Variables (patients/knees) 381/483
Patient Factor Sex (Male 63/77, Female 318/406) 18 22/79 88
Age (years) 74 [69, 78]
Body weight (kg), Male/Female 58 [53,66]
Body height (cm) 151 [147, 155]
Body mass index (kg/m2), Male/Female 26 [23, 28]
HSS score 45 [37, 51]
Comorbidity
Hypertension (Yes/No) 301/182
Hyperlipidemia (Yes/No) 95/388
Diabetes Mellitus (Yes/No) 59/424
Smoking history (Yes/No) 24/459
Surgical Factor Operative Time (min.) 55 [50, 62]
Torniquet Time (min.) 58 [53, 65]
Total blood loss (ml) 430 [200, 720]
Open in a new tab

Data are presented as n or median [25th, 75th percentile]. HSS, Hospital for Special Surgery 19.

2.1. Statistical analysis

Because some variables did not show a normal distribution with the Kolmogorov–Smirnov statistic, the Shapiro–Wilk test, or the Q–Q plot, these variables were analyzed using a nonparametric method. All variables are expressed as medians (25th and 75th percentiles). Differences between two groups were compared using the Wilcoxon signed-rank test and the Mann–Whitney U test. Differences in dependent measures among three groups were analyzed using the Friedman test and post hoc multiple comparisons using Scheffé’s method. Spearman's rank sum test was used to assess correlation between two variables. The strength of the correlation between the rank coefficients was defined as strong (0.70–1.0), moderate (0.40–0.69), or weak (0.20–0.39). Regression analysis was performed to examine the relationship between two continuous variables. Multiple regression analysis was performed to identify the variables that were significantly associated with SA and ΔSA with a stepwise selection method employed to select significant variables. All analyses were performed using the IBM SPSS Statistics ver. 23 (IBM Japan, Tokyo, Japan) and R version 4.3.0 (GNU) with the related packages. In all tests, a p-value <0.05 was considered statistically significant.

3. Results

Median values for SA0, SA1W, and SA4W were 4.4 (4.2, 4.6) g/dL, 3.8 (3.6, 3.9) g/dL, and 4.2 (4.0, 4.3) g/dL, respectively, which was significantly different between groups (p < 0.001). The distribution of patients across SA concentration ranges (<3.0, 3.0–3.4, 3.5–3.9, 4.0–4.4, and ≥4.5 g/dL) is shown in Table 2; the prevalence of low SA (<3.5 g/dL) preoperatively, 1 week postoperatively and 4 weeks postoperatively, was 0.8 %, 13 %, and 1.8 %, respectively. The median changes in SA between timepoints were as follows: ΔSA1W−0 −13.7 % (−17.7 %, −9.5 %), ΔSA4W–1W 9.6 % (5.7 %, 13.1 %), and ΔSA4W−0 −4.5 % (−8.2 %, −0.2 %); these changes were significantly different between groups (p < 0.001). The distribution of patients across SA concentration change ranges (≤−20 %, −20 < SA ≤ −10 %, −10 < ΔSA ≤0 %, and >0 %) is shown in Table 3.

Table 2.

Distribution of patients between serum albumin concentration ranges at each time point.

Serum Albumin Level (g/dL)
 Number of Patients (%)
N (%) Pre-op* 1 week post-op* 4 weeks post-op*
Median [quasi] 4.4 [4.2, 4.6] 3.8 [3.6, 3.9] 4.2 [4.0, 4.3]
<3.0 2 (0.4) 3 (0.6) 1 (0.2)
3.0–3.4 2 (0.4) 60 (12.4) 8 (1.7)
3.5–3.9 38 (7.9) 301 (62.3) 93 (19.3)
4.0–4.4 251 (52.0) 117 (24.2) 310 (64.2)
≥4.5 190 (39.3) 2 (0.4) 71 (14.7)
Open in a new tab

*p < 0.001 (among three groups) using Friedman test, and each group (p < 0.001) using Scheffé’s method.

Table 3.

Distribution of patients between serum albumin (SA) concentration change ranges expressed as a percentage of preoperative SA.

Period; N (%) ΔAlb ≤ −20 % −20 < ΔAlb ≤ −10 % −10 < ΔAlb ≤0 % 0 % < ΔAlb
(Po 1W- Pre)/Pre −13.7 (−17.7, −9.5)* 58 (12.0) 291 (60.2) 126 (26.1) 8 (1.7)
(Po4w-Po1w)/Pre 9.6 (5.7, 13.1)* 0 (0) 1 (0.2) 33 (6.8) 449 (93.0)
(Po 4w- Pre)/Pre −4.5 (−8.2, −0.2)* 5 (1.0) 80 (16.6) 284 (58.8) 114 (23.6)
Open in a new tab

*p < 0.001 (among 3 groups) by the Friedman's test, and each group (p < 0.001) with Scheffé’s method.

Correlations between SA, ΔSA, and each variable factor are shown in Table 4. SA0, SA1W and SA4W were significantly positively correlated: SA0 vs. SA1W; r = 0.539, p < 0.001; SA0 vs. SA4W; r = 0.513, p < 0.001; and SA1W vs. SA4W; r = 0.573, p < 0.001. The regression equations for SA1W and SA4W vs. SA0 were 1.355 + 0.552 × SA0 and 1.874 + 0.526 × SA0, respectively.

Table 4.

Correlations between two variables using Spearman's correlation coefficient.

Variables SA0 SA1w SA4w ΔSA1W-0 ΔSA4W-1W ΔSA 4W-0 Age BMI HSS Operative Time Tourniquet Time TB Loss
SA0 1.0
SA1w 0.539 < 0.001 1.0
SA4w 0.513 < 0.001 0.573 < 0.001 1.0
ΔSA1W-0 −0.329 < 0.001 0.554 < 0.001 0.146 0.001 1.0
ΔSA4W-1W −0.124 0.006 −0.470 < 0.001 0.375 < 0.001 −0.387 < 0.001 1.0
ΔSA4W-0 −0.413 < 0.001 0.056 0.220 0.491 < 0.001 0.510 < 0.001 0.537 < 0.001 1.0
Age −0.209 < 0.001 −0.231 < 0.001 −0.277 < 0.001 −0.066 0.150 −0.009 0.836 −0.079 0.082 1.0
BMI 0.063 0.165 0.140 0.002 0.092 0.044 0.059 0.192 −0.049 0.282 0.037 0.412 −0.206 < 0.001 1.0
HSS 0.078 0.085 0.066 0.150 0.119 0.009 0.008 0.862 0.052 0.251 0.038 0.408 −0.183 < 0.001 0.025 0.577 1.0
Operative Time −0.004 0.938 −0.113 0.013 −0.088 0.055 −0.103 0.024 0.031 0.502 −0.079 0.085 −0.024 0.594 0.078 0.085 −0.041 0.372 1.0
Tourniquet Time −0.005 0.915 −0.104 0.023 −0.108 0.018 −0.105 0.021 −0.008 0.861 −0.106 0.020 −0.001 0.975 0.023 0.614 −0.027 0.559 0.845 < 0.001 1.0
TB Loss −0.069 0.130 −0.118 0.010 0.023 0.608 −0.066 0.148 0.153 0.001 0.093 0.040 −0.040 0.382 −0.034 0.450 −0.002 0.963 −0.110 0.016 −0.024 0.59219 1.0
Open in a new tab

SA0, serum albumin (SA) values preoperatively; SA1W, SA values 1 week postoperatively; SA4W, SA values 4 weeks postoperatively; ΔSA1W−0, the difference between SA1W and SA0; ΔSA4W–1W, the difference between SA4W and SA1W; ΔSA4W−0, the difference between SA4W and SA0; BMI, body mass index; HSS, Hospital for Special Surgery 19; TB, total blood. The upper row shows the correlation coefficient, and the bottom shows the p-value. Bold numbers indicate a statistically significant correlation (p < 0.05, r ≥ 0.20 or ≤ −20).

ΔSA1W−0 was significantly negatively correlated with ΔSA4W–1W (r = −0.387, p < 0.001). ΔSA4W−0 was significantly positively correlated with ΔSA1W−0 (r = 0.510, p < 0.001) and ΔSA4W–1W (r = 0.537, p < 0.0001). SA concentration at each timepoint showed a weak but significant negative correlation with age: SA0; r = −0.209, p < 0.001; SA1W; r = −0.239, p < 0.001; and SA4W: r = −0.277, p < 0.001. In addition, there were significant correlations between BMI and age (r = −0.206, p < 0.001), and tourniquet time and operative time (r = 0.845, p < 0.001).

The results of the multiple regression analysis are presented in Table 5a–f. SA concentration at each timepoint was significantly associated with the other SA timepoint measures and the ΔSA between corresponding measurement timepoints (Table 5a–c). Likewise, each ΔSA measure showed a significant relationship with SA at corresponding measurement timepoints (Table 5d–f). ΔSA1W−0 was significantly associated with ΔSA4W–1W, while ΔSA4W−0 was significantly related to ΔSA1W−0 and ΔSA4W–1W (Table 5d–f). Age was significantly associated with SA1W and SA4W, but not SA0.

Table 5.

Results of the multiple regression analysis using a stepwise variable selection method.

a. SA0
B S.E. β Sig. 95%CI
SA1w 0.547 0.031 0.514 <0.001 0.486 0.608
ΔSA1w-0 −0.558 0.031 −0.483 <0.001 −0.619 −0.496
SA4w 0.447 0.031 0.445 <0.001 0.387 0.508
ΔSA4w-0 −0.443 0.031 −0.422 <0.001 −0.505 −0.382
Intercept 0.015 0.029 0.619 −0.043 0.072
b. SA1w
B S.E. β Sig. 95%CI
SA4w 0.431 0.031 0.456 <0.001 0.369 0.492
ΔSA4w-1w −0.438 0.031 −0.394 <0.001 −0.499 −0.376
SA0 0.565 0.031 0.601 <0.001 0.504 0.626
ΔSA4w-0 0.560 0.032 0.517 <0.001 0.497 0.623
Age 0.001 0.000 0.022 0.001 0.000 0.001
intercept −0.041 0.039 0.298 −0.117 0.036
c. SA4w
B S.E. β Sig. 95%CI
SA1w 0.486 0.036 0.459 <0.001 0.415 0.557
ΔSA4w-1w 0.481 0.036 0.409 <0.001 0.410 0.553
Age −0.001 0.000 −0.022 0.001 −0.001 0.000
ΔSA4w-0 0.512 0.036 0.491 <0.001 0.441 0.583
SA0 0.498 0.036 0.500 <0.001 0.427 0.569
intercept 0.136 0.042 0.001 0.053 0.219
d. ΔSA1w-0
B S.E. β Sig. Exp(β)
SA0 −0.51 0.033 −0.587 <0.001 −0.574 −0.446
SA1w 0.512 0.033 0.554 <0.001 0.448 0.576
ΔSA4w-1w −0.465 0.033 −0.452 <0.001 −0.529 −0.400
ΔSA4w-0 0.473 0.032 0.519 <0.001 0.409 0.537
Intercept −0.029 0.029 0.314 −0.085 0.028
e. ΔSA4w-1w
B S.E. β Sig. Exp(β)
SA1w −0.447 0.032 −0.496 <0.001 −0.51 −0.384
SA4w 0.448 0.032 0.527 <0.001 0.386 0.511
ΔSA1w-0 −0.545 0.032 −0.56 <0.001 −0.608 −0.481
ΔSA4w-0 0.534 0.032 0.603 <0.001 0.471 0.598
Intercept −0.007 0.03 0.820 −0.066 0.052
f. ΔSA4w-0
B S.E. β Sig. Exp(β)
SA0 −0.463 0.032 −0.486 <0.001 −0.525 −0.401
SA4w 0.462 0.032 0.481 <0.001 0.399 0.524
ΔSA1w-0 0.526 0.032 0.479 <0.001 0.462 0.59
ΔSA4w-1w 0.526 0.032 0.467 <0.001 0.464 0.589
Intercept −0.463 0.032 0.835 −0.052 0.065
Open in a new tab

SA and ΔSA were not any significantly correlations with discrete study variable groups (sex; presence of hypertension, hyperlipidemia, and diabetes mellitus; and history of smoking) pre- or postoperatively (Table 6, Table 7).

Table 6.

Differences in serum albumin concentration (g/dL) according to discrete study variables pre- and postoperatively.

Variables (Knees)
 Median (Interquartile)
 p
 Median (Interquartile)
 p
 Median (Interquartile
 p
Period Preoperative SA 1 week postoperative SA 4 weeks postoperative SA
Sex: Male (77)/Female (406) 4.3 [4.2, 4.5]/4.3 [4.2,4.6] 0.384 3.7 [3.5, 4.0]/3.7 [3.6,3.9] 0.392 4.2 [4.0, 4.4]/4.0 [4.0, 4.3] 0.822
Hypertension: yes (302)/no (181) 4.4 [4.2,4.6]/4.3 [4.1,4.5] 0.139 3.8 [3.6,4.0]/3.7 [3.5,3.9] 0.179 4.2 [4.0,4.3]/4.1 [4.0,4.3] 0.962
Hyperlipidemia: yes (95)/no (388) 4.4 [4.2,4.6]/4.3 [4.2,4.6] 0.830 3.8 [3.6,4.0]/3.7 [3.6,3.9] 0.171 4.2 [4.0,4.3]/4.2 [4.0,4.3] 0.946
Diabetes Mellites: yes (59)/no (424) 4.4 [4.2,4.6]/4.4 [4.2,4.6] 0.603 3.8 [3.6,3.9]/3.7 [3.6,3.9] 0.406 4.2 [4.0,4.4]/4.2 [4.0,4.3] 0.143
Smoking history: yes (24)/no (459) 4.3 [4.2,4.5]/4.4 [4.2,4.6] 0.319 3.7 [3.5,3.9]/3.8 [3.6,3.9] 0.740 4.2 [4.0,4.5]/4.2 [4.0,4.3] 0.371
Open in a new tab

Data are presented as n or median [25th percentile, 75th percentile] and were analyzed using the Wilcoxon rank sum test.

Table 7.

Differences in serum albumin concentration changes (%) according to discrete study variables pre- and postoperatively.

Variables (Knees)
 Median (Interquartile)
 p
 Median (Interquartile)
 p
 Median (Interquartile
 p
Period ΔSA1W-0 ΔSA4W-1W ΔSA4W-0
Sex: Male (77)/Female (406) −13.6 [-17.8, −8.4] −17.0 [-17.7, −9.6] 0.823 10.5 [6.9, 14.0] 10.0 [5.5, 12.8] 0.526 −3.0 [-6.7, 0.4] −4.7 [-8.4, −0.4] 0.637
Hypertension: yes (302)/no (181) −13.7 [-17.6, −9.4] −13.5 [-18.2, −9.7] 0.586 8.9 [4.9,12.8] 10.1 [6.7,13.5] 0.561 −4.8 [-8.4, −0.9] −4.1 [-7.8, 0.9] 0.335
Hyperlipidemia: yes (95)/no (388) −13.2 [-16.7, −8.9] −13.9 [-17.9, −9.7] 0.161 8.9 [5.0,12.5] 9.7 [6.0,13.3] 0.073 −5.0 [-8.4, −0.0] −4.3 [-8.1, −0.3] 0.424
Diabetes Mellites: yes (59)/no (424) −13.6 [-18.3, −9.9] −13.8 [-17.7, −9.3] 0.548 9.6 [6.1,13.6] 9.6 [5.6,13.0] 0.400 −4.3 [-8.2, −0.9] −4.5 [-8.2, −0.1] 0.284
Smoking history: yes (24)/no (459) −13.7 [-19.2, −9.2] −13.7 [-17.7, −9.5] 0.775 13.7 [8.1,16.0] 9.5 [5.6,12.8] 0.520 −1.5 [-3.7, 2.8] −4.7 [-8.4, −0.3] 0.607
Open in a new tab

Data are presented as n or median [25th percentile, 75th percentile] and were analyzed using the Wilcoxon rank sum test.

4. Discussion

This study aimed to investigate the perioperative dynamics of SA and ΔSA in TKA patients and identify any patient or surgical factors affecting SA concentration. We identified three important results. First, correlation analysis suggests that SA0 concentration influenced both SA1W and SA4W concentration. Second, the ΔSA results showed that an initial 13.7 % decrease in SA observed 1 week postoperatively recovered to 95.5 % of preoperative SA 4 weeks postoperatively; given the negative correlation between ΔSA1W−0 and ΔSA4W–1W, a greater decrease in SA 1 week postoperatively was associated with a greater recovery at 4 weeks postoperatively. Finally, results from the multiple regression analysis suggest that postoperative SA concentrations were affected by age.

The present study revealed a significant correlation between SA0 and both SA1W and SA4W. To date, TKA-related SA studies have mainly used preoperative values,1,2,4, 5, 6,9,16,21 although some16,22, 23, 24, 25, 26 have reported postoperative SA values. However, of these, the exact outcomes measured varied: while some studies compared the effect of preoperative amino acid (AA) administration on postoperative clinical outcomes,22,23 others compared muscle mass24, 25, 26 between the treated and non-treated groups, and another reported SA 3 days postoperatively.16 However, none of these studies focused on the correlations we observed, nor were they mentioned. No detailed studies of perioperative SA trends or factors influencing SA after TKA have been conducted; therefore, additional studies are required to confirm the associations we noted between pre- and postoperative SA.

With regards to the threshold for perioperative low SA in TKA, many reports1,5,6,8, 9, 10,13,14,16,26, 27, 28 use a definition of <3.5 g/dL. Across studies, the proportion of patients meeting this criterion ranges from 2.7 % to 9.9 %5,9,10,16; however, in the present study, the frequency was less than 1 % preoperatively and 2 % 4 weeks postoperatively. In addition, cutoff values used to define malnutrition (i.e., low SA) are mixed, ranging from 3.0 16 to 3.94 g/dL,12 depending on the study. Black et al.12 suggested that a cutoff value of 3.5 g/dL for albumin level after TKA may still miss high-risk patients; in a univariate model of 90-day readmission, these authors therefore concluded that the optimal albumin cutoff was 3.94 g/dL. In seven large, multicenter studies, Mobagwu et al.29 reported that the trend for postoperative complications was the same regardless of whether the cutoff was 3.0 g/dL, 3.5 g/dL, or 3.9 g/dL. Likewise, Morey et al.21 reported no association between postoperative wound-related complications and malnutrition defined using various SA cutoff values. Thus, the authors concluded that the accuracy and reliability of these markers, which are commonly used as serological surrogate markers of malnutrition for postoperative wound complications in TKA patients, was highly questionable. These studies emphasize the need for easily detectable, reliable surrogate markers of malnutrition, in addition to the previously reported SA0 assessments.

As a result of the low reliability of a single, preoperative SA measure, ΔSA was used as an additional surrogate marker in this study to better understand the effect of the entire perioperative SA profile. Several studies have reported ΔSA before and after TKA23,24 when investigating the effects of preoperative AA administration, but not to characterize SA dynamics after TKA surgery. In an AA supplementation study, Dreyer et al.24 found that the mean SA0 level in the untreated group was 4.05 ± 0.06 g/dL, and remained at 93.6 %, 79.3 %, 92.8 %, and 97.3 % of this value on the day of surgery, 2 days, 2 weeks, and 6 weeks later, respectively. In the AA-treated group, the mean SA0 level was 4.02 ± 0.05 g/dL and remained at 95.3 %, 78.1 %, 92.8 %, and 98.0 %, respectively. In both groups, SA values were at their lowest two days postoperatively and tended to improve by two weeks postoperatively. Similarly, Ueyama et al.24 reported SA trends to determine the efficacy of essential AA administration (9 g daily) for preventing muscle atrophy after TKA surgery. The authors reported a mean SA0 of 4.4 g/dL (range, 3.3–5.0 g/dL) in the treated group and 4.5 g/dL (3.7–5.1 g/dL) in the untreated group; the lowest values in both groups were more than 10 % lower 3 weeks postoperatively, but 95 % and 89 % recovery 4 weeks postoperatively, respectively. Overall, the reported trends in these studies are similar to our results and show that perioperative SA values are highly dynamic, meaning evaluation of multiple SA values is important in any field of surgery. Indeed, ΔSA is superior to SA0 in its ability to identify patients with postoperative complications in liver surgery.30

In the present study, the significant negative correlation between ΔSA1W−0 and ΔSA4W–1W suggests that a larger decrease in SA 1 week postoperatively was associated with a better recovery 4 weeks after TKA, indicating good SA recovery in TKA patients. Therefore, to obtain a good ΔSA4W−0, the administration of essential AAs could be considered as a measure to decrease SA up to one week postoperatively to subsequently accelerate SA increase from 1 to 4 weeks postoperatively. In a systematic review of TJA, Burgess et al.21 found that preoperative optimization of nutritional status, including SA, is particularly beneficial for undernourished, frail, or elderly patients, which may help in the management of surgical stress reactions. We are currently analyzing whether and how ΔSA can be used in clinical practice to guide clinicians in perioperative management of TKA patients.

Factors that influence SA in the perioperative period of TKA include sex,4,5,8,14 race,5,8,14 older age,4,5,8,14 American Society of Anesthesiologists31 grade,5,10,14 BMI,8,14 general anesthesia compared to regional, spinal, epidural and other,5 smoking,16 alcohol consumption,16 preoperative albumin,16 intraoperative blood loss,16 and operative time.16 Previously, Li et al.16 reported significant differences between patients in SA0 <3.0 g/dL and ≥3.0 g/dL groups in postoperative SA, sex, smoking, alcohol consumption, preoperative albumin, intraoperative blood loss, and operative time. Furthermore, they reported that in patients who underwent TKA, SA0, operative time, and intraoperative blood loss were significant independent risk factors for the need for postoperative albumin supplementation. Kishawi et al.8 found that belonging to the SA0 <3.5 g/dL group was associated with age >60 years, female gender, African American ethnicity, and the presence of obesity and comorbidities. However, in the present study, multiple regression analysis showed that only SA0 values significantly influenced both postoperative SA and ΔSA, and only age significantly influenced postoperative SA. Although no conclusive reason can be given from the data in the current study, it may be because all TKA procedures were performed by the same surgeon using the same technique and prosthesis design, so surgeon-derived bias was less likely. In addition, patient-derived bias was less likely because patients were of the same race, older age (which may have less gender-based bias), and complications were well controlled. Nonetheless, our results suggest that special attention should be paid to optimizing SA in elderly patients with low SA0.

This study has several limitations. The most important limitation is that the study design was retrospective and observational. Therefore, prospective studies are required to confirm the results of this study. The second limitation is that not all nutritional parameters were measured in patients in the current study; therefore, only SA levels were assessed. The third limitation is that the lack of clinical relevance and suggested cutoff values of SA. We are currently analyzing the cutoff values of SA and whether and how ΔSA can be used in clinical practice to guide clinicians in perioperative management of TKA patients. Finally, this was a single-center, single-race study; a multicenter, multiracial setting is needed to confirm the universality of the results. The main strength of this study is that it is the first to identify SA trends during the perioperative period, specifically, before, 1 week after, and 4 weeks after TKA. We believe that the results of this study can be used to help determine how to effectively counteract low SA levels in the perioperative period.

In conclusion, overall, we found the incidence of low preoperative SA (<3.5 g/dL) in TKA patients was less than 1 %. The average ΔSA was −13.7 % at 1 week postoperatively, recovering to 95.5 % of preoperative values by 4 weeks. Correlation analysis showed that a greater degree of SA decline at 1 week postoperatively was associated with a greater degree of recovery at 4 weeks postoperatively, with less than 2 % of patients having low SA4W (<3.5 g/dL). Factors affecting SA dynamics in the perioperative period were SA0 and age. Given the low prevalence of low pre- and postoperative SA and the degree of recovery, conventional cutoff values, and preventive measures for low SA after TKA may need to be reconsidered.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding

None declared.

Ethical review committee statement

The local institutional review board approved this study. All patients provided informed consent.

Approval for this study was obtained from the Research Board of Healthcare Corporation Ashinokai, Gyoda, Saitama, Japan [ID number: 2020-1].

CRediT authorship contribution statement

Yoshinori Ishii: contributed to the study, Conceptualization, and design, Writing – original draft, the article, and ensured the accuracy of the, Data curation, and, Formal analysis. Hideo Noguchi: contributed to the study, Conceptualization, and design and to the, Formal analysis, and interpretation of the, Data curation. Junko Sato: contributed to the study, Conceptualization, and design and to the, Formal analysis, and interpretation of the, Data curation. Ikuko Takahashi: contributed to the study, Conceptualization, and design and to the, Formal analysis, and interpretation of the, Data curation. Hana Ishii: contributed to the, Data curation, collection. Ryo Ishii: contributed to the, Data curation, collection. Kai Ishii: contributed to the, Data curation, collection. Shin-ichi Toyabe: provided statistical expertise and contributed to ensuring the accuracy of the, Data curation, and, Formal analysis, All authors approved the final manuscript.

Declaration of competing interest

The authors declare that they have no competing interests.

Acknowledgment

We thank Syohei Yoshizawa RN for his contributions in gathering the data, and Edanz Group (https://en-author-services.edanz.com/ac), for editing a draft of this manuscript.

Footnotes

☆

The present work was performed at the Ishii Orthopaedic and Rehabilitation Clinic, 1089 Shimo-Oshi, Gyoda, Saitama 361-0037, Japan.

Contributor Information

Yoshinori Ishii, Email: ishii@sakitama.or.jp.

Hideo Noguchi, Email: hid_166super@mac.com.

Junko Sato, Email: jun-sato@hotmail.co.jp.

Ikuko Takahashi, Email: itakahashi110@gmail.com.

Hana Ishii, Email: hanamed12@gmail.com.

Ryo Ishii, Email: kmuyakyu@gmail.com.

Kai Ishii, Email: kai.nd1209@live.com.

Shin-ichi Toyabe, Email: toyabe@med.niigata-u.ac.jp.

References

  • 1.Blevins K., Aalirezaie A., Shohat N., Parvizi J. Malnutrition and the development of periprosthetic joint infection in patients undergoing primary elective total joint arthroplasty. J Arthroplasty. 2018;33:2971–2975. doi: 10.1016/j.arth.2018.04.027. [DOI] [PubMed] [Google Scholar]
  • 2.Huang R., Greenky M., Kerr G.J., Austin M.S., Parvizi J. The effect of malnutrition on patients undergoing elective joint arthroplasty. J Arthroplasty. 2013;28(8 suppl):21–24. doi: 10.1016/j.arth.2013.05.038. [DOI] [PubMed] [Google Scholar]
  • 3.Phillips J.L.H., Ennis H.E., Jennings J.M., Dennis D.A. Screening and management of malnutrition in total joint arthroplasty. J Am Acad Orthop Surg. 2023;31:319–325. doi: 10.5435/JAAOS-D-22-01035. [DOI] [PubMed] [Google Scholar]
  • 4.Bala A., Ivanov D.V., Huddleston J.I., Goodman S.B., Maloney W.J., Amanatullah O.F. The cost of malnutrition in total joint arthroplasty. J Arthroplasty. 2020;35:926–932. doi: 10.1016/j.arth.2019.11.018. [DOI] [PubMed] [Google Scholar]
  • 5.Fu M.C., McLawhorn A.S., Padgett D.E., Cross M.B. Hypoalbuminemia is a better predictor than obesity of complications after total knee arthroplasty: a propensity score-adjusted observational analysis. HSS J. 2017;13:66–74. doi: 10.1007/s11420-016-9518-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Dubé M.D., Rothfusz C.A., Emara A.K., et al. Nutritional assessment and interventions in elective hip and knee arthroplasty: a detailed review and guide to management. Curr Rev Musculoskelet Med. 2022;15:311–322. doi: 10.1007/s12178-022-09762-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Greene K.A., Wilde A.H., Stulberg B.N. Preoperative nutritional status of total joint patients. Relationship to postoperative wound complications. J Arthroplasty. 1991;6:321–325. doi: 10.1016/s0883-5403(06)80183-x. [DOI] [PubMed] [Google Scholar]
  • 8.Kishawi D., Schwarzman G., Mejia A., Hussain A.K., Gonzalez M.H. Low preoperative albumin levels predict adverse outcomes after total joint arthroplasty. J. Bone Joint Surg. Am. 2020;102:889–895. doi: 10.2106/JBJS.19.00511. [DOI] [PubMed] [Google Scholar]
  • 9.Man S.L.C., Chau W.W., Chung K.Y., Ho K.K.W. Hypoalbuminemia and obesity class I are reliable predictor of peri-prosthetic joint infection in patient undergoing elective total knee arthroplasty. Knee Surg Relat Res. 2020;32:21. doi: 10.1186/s43019-020-00040-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Nelson C.L., Elkassabany N.M., Kamath A.F., Liu J. Low albumin levels, more than morbid obesity, are associated with complications after TKA. Clin Orthop Relat Res. 2015;473:3163–3172. doi: 10.1007/s11999-015-4333-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Klingenstein G.G., Schoifet S.D., Jain R.K., Reid J.J., Porat M.D., Otegbeye M.K. Rapid discharge to home after total knee arthroplasty is safe in eligible medicare patients. J Arthroplasty. 2017;32:3308–3313. doi: 10.1016/j.arth.2017.06.034. [DOI] [PubMed] [Google Scholar]
  • 12.Black C.S., Goltz D.E., Ryan S.P., et al. The role of malnutrition in ninety-day outcomes after total joint arthroplasty. J Arthroplasty. 2019;34:2594–2600. doi: 10.1016/j.arth.2019.05.060. [DOI] [PubMed] [Google Scholar]
  • 13.O'Connor M.I., Bernstein J., Huff T. Movement is life-Optimizing patient access to total joint arthroplasty: malnutrition disparities. J Am Acad Orthop Surg. 2022;30:1007–1010. doi: 10.5435/JAAOS-D-21-00415. [DOI] [PubMed] [Google Scholar]
  • 14.Holbert S.E., Brennan J.C., Johnson A.H., Turcotte J.J., King P.J., MacDonald J.H. The effects of hypoalbuminemia in obese patients undergoing total joint arthroplasty. Arch Orthop Trauma Surg. 2023;143:4813–4819. doi: 10.1007/s00402-023-04786-1. [DOI] [PubMed] [Google Scholar]
  • 15.Torchia M.T., Khan I.A., Christensen D.D., Moschetti W.E., Fillingham Y.A. Universal screening for malnutrition prior to total knee arthroplasty is cost-effective: a Markov analysis. J Arthroplasty. 2023;38:443–449. doi: 10.1016/j.arth.2022.10.014. [DOI] [PubMed] [Google Scholar]
  • 16.Li A.A., Zhang Y., Zhang H., et al. The role of routine laboratory tests after unilateral total knee arthroplasty. BMC Muscoskel Disord. 2022;23:564. doi: 10.1186/s12891-022-05509-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Martin J.R., Jennings J.M., Dennis D.A. Morbid obesity and total knee arthroplasty: a growing problem. Am Acad Orthop Surg. 2017;25:188–194. doi: 10.5435/JAAOS-D-15-00684. [DOI] [PubMed] [Google Scholar]
  • 18.Schneider A.M., Brown N.M. Should insulin-dependent diabetic patients be screened for malnutrition before total joint arthroplasty? A cohort at risk. J Am Acad Orthop Surg. 2021;29:673–680. doi: 10.5435/JAAOS-D-20-00729. [DOI] [PubMed] [Google Scholar]
  • 19.Alicea J. In: third ed. Insall J.N., Scott W.N., editors. vol. 2. Churchill Livingstone; New York: 2001. Scoring systems and their validation for the arthritic knee; pp. 1507–1515. (Surgery of the Knee). [Google Scholar]
  • 20.Ishii Y., Noguchi H., Takeda M., Sato J., Ezawa N., Toyabe S. Changes in lower extremity 3-dimensional load-bearing axes before and after mobile-bearing total knee arthroplasty. J Arthroplasty. 2012;27:1203–1209. doi: 10.1016/j.arth.2011.12.013. [DOI] [PubMed] [Google Scholar]
  • 21.Burgess L.C., Phillips S.M., Wainwright T.W. What is the role of nutritional supplements in support of total hip replacement and total knee replacement surgeries? A systematic review. Nutrients. 2018;10:820. doi: 10.3390/nu10070820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Schroer W.C., LeMarr A.R., Mills K., Childress A.L., Morton D.J., Reedy M.E. Chitranjan S. Ranawat Award: elective joint arthroplasty outcomes improve in malnourished patients with nutritional intervention: a prospective population analysis demonstrates a modifiable risk factor. Bone Joint Lett J. 2019;101-B(7_Supple_C):17–21. doi: 10.1302/0301-620X.101B7.BJJ-2018-1510.R1. 2019. [DOI] [PubMed] [Google Scholar]
  • 23.Dreyer H.C., Owen E.C., Strycker L.A., et al. Essential amino acid supplementation mitigates muscle atrophy after total knee arthroplasty: a randomized, double-blind, placebo-controlled trial. JBJS Open Access. 2018;3 doi: 10.2106/JBJS.OA.18.00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ueyama H., Kanemoto N., Minoda Y., Taniguchi Y., Nakamura H. Chitranjan S. Ranawat Award: perioperative essential amino acid supplementation suppresses rectus femoris muscle atrophy and accelerates early functional recovery following total knee arthroplasty. Bone Joint Lett J. 2020;102-B(6_Supple_A):10–18. doi: 10.1302/0301-620X.102B6.BJJ-2019-1370.R1. 2020. [DOI] [PubMed] [Google Scholar]
  • 25.Ueyama H., Kanemoto N., Minoda Y., Taniguchi Y., Nakamura H. Perioperative essential amino acid supplementation facilitates quadriceps muscle strength and volume recovery after TKA: a double-blinded randomized controlled trial. J. Bone Joint Surg. Am. 2023;105:345–353. doi: 10.2106/JBJS.22.00675. [DOI] [PubMed] [Google Scholar]
  • 26.Kamath A.F., Nelson C.L., Elkassabany N., Guo Z., Liu J. Low albumin is a risk factor for complications after revision total knee arthroplasty. J Knee Surg. 2017;30:269–275. doi: 10.1055/s-0036-1584575. [DOI] [PubMed] [Google Scholar]
  • 27.Rahman T.M., Fleifel D., Padela M.T., et al. Interventions for obesity and nutritional status in arthroplasty patients. JBJS Rev. 2020;8 doi: 10.2106/JBJS.RVW.19.00161. [DOI] [PubMed] [Google Scholar]
  • 28.Morey V.M., Song Y.D., Whang J.S., Kang Y.G., Kim T.K. Can serum albumin level and total lymphocyte count be surrogates for malnutrition to predict wound complications after total knee arthroplasty? J Arthroplasty. 2016;31:1317–1321. doi: 10.1016/j.arth.2015.12.004. [DOI] [PubMed] [Google Scholar]
  • 29.Mbagwu C., Sloan M., Neuwirth A.L., et al. Preoperative albumin, transferrin, and total lymphocyte count as risk markers for postoperative complications after total joint arthroplasty: a systematic review. J Am Acad Orthop Surg Glob Res Rev. 2020;4 doi: 10.5435/JAAOSGlobal-D-19-00057. e19.00057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Labgaa I., Cano L., Mangana O., et al. An algorithm based on the postoperative decrease of albumin (ΔAlb) to anticipate complications after liver surgery. Perioperat Med. 2022;11:53. doi: 10.1186/s13741-022-00285-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Anonymous. American society of Anaesthesiologists physical status classification system. http://www.asahq.org/resources/clinical-information/asa-physical-statusclassification-system [DOI] [PMC free article] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Articles from Journal of Orthopaedics are provided here courtesy of Elsevier

RESOURCES