A pilot observational study measuring acute sarcopenia in older colorectal surgery patients

Objective To explore variability in acute changes in muscle mass and function in older patients undergoing elective colorectal surgery, as well as feasibility of measures, in order to refine study processes to inform the protocol for a larger study. Results Results are presented for seven participants recruited to this pilot study. It is possible to perform serial measurements of bilateral anterior thigh thickness (BATT) and handgrip strength prior to, within 24 h of surgery, and 1 week postoperatively. Gait speed can be reliably measured preoperatively and at 1 week postoperatively. In this pilot study, BATT and gait speed declined at 1 week postoperatively (median BATT 4.17 cm, 3.47 cm, p = 0.028; median gait speed 0.89 m/s, 0.83 m/s, p = 0.043). Baseline hsCRP correlated with change in BATT (τb = 0.73, p = 0.04) and baseline DHEA-S correlated with change in gait speed (τb = 0.87, p = 0.02). This pilot study has assisted to refine the protocol for our larger study, which will further characterise these changes. Electronic supplementary material The online version of this article (10.1186/s13104-019-4049-y) contains supplementary material, which is available to authorized users.


Introduction
Sarcopenia is defined as reduced muscle mass and function [1]. European Working Group on Sarcopenia in Older People 2 (EWGSOP2) distinguishes acute from chronic sarcopenia [2]. Acute sarcopenia develops within 6 months, normally triggered by a stressor event [2]. Hospitalisation is associated with bedrest and inflammatory response, which can drive catabolism/reduce anabolism, inducing reductions in muscle mass and function; this is exacerbated by age [3,4]. Precise mechanisms remain unclear. Overview of acute sarcopenia including proposed mechanisms and effectors has been published elsewhere [5].
Colorectal surgery is commonly performed on older adults for localised cancer resection [6,7] and triggers acute inflammatory response [8]. Patients do not normally experience preoperative functional decline or cachexia [9]. Characterising acute sarcopenia in these patients will help to increase understanding, identify mechanistic pathways, and target interventions, with potential to improve postoperative recovery and longerterm physical function. This study aimed to refine the protocol for a larger study by exploring feasibility and variability of measures.

Study design
Participants were recruited from Queen Elizabeth Hospital Birmingham preoperative assessment clinic during August-October 2017. We included people aged 65 and older, planned to undergo major colorectal surgery (any indication), and able to give informed consent. Exclusion criteria were life expectancy 30 days or less and inability to understand verbal English. Known muscular disorders were not an exclusion criterion. Measurements were taken prior to admission (Visit 1), within 24 h of surgery  (Visit 2), and 1 week postoperatively or day of discharge (Visit 3). A single geriatrician with training in procedures performed all assessments (CW). All assessments not possible to perform and reasons for this were recorded.

Muscle mass measurement
Quadriceps muscle thickness was measured using B-mode ultrasonography (Venue 50, GE Healthcare, UK). Participants were positioned semi-upright, with a firm wedge below the knees, and advised to relax their muscles. Measurements were taken in transverse plane at mid-point between greater trochanter and knee joint line. Thickness of subcutaneous tissue (SC), rectus femoris (RF), and vastus intermedius (VI), excluding fascia, were measured. Three measurements were taken on each side at each visit and means calculated; a fourth was taken if measurements differed more than 10%. Total bilateral anterior thigh thickness (BATT) (right RF + right VI + left RF + left VI), and BATT:SC ratio (BATT-SCR) were calculated. This method has excellent intra-rater and inter-rater reliability and mean BATT correlates with bioelectrical impedance analysis (BIA) in healthy young adults [10].

Muscle strength measurement
Handgrip strength was measured at each visit using a Jamar hydraulic handheld dynamometer (Patterson Medical, Warrenville, IL, USA). Participants were positioned with elbows flexed at 90°, forearms in neutral position and advised to "squeeze as hard as [they] can". Highest recording of two measurements on each side was used for analysis [11].

Physical performance testing
Short Physical Performance Battery (SPPB) was scored at Visit 1; incorporates gait speed, side-by-side/tandem stand, and five chair stands [12]. We initially planned to measure SPPB at all visits; upon consideration of effect of intra-abdominal strain limiting chair stands, gait speed alone was measured at Visit 3. Gait speed is a widely accepted measure of physical performance [1,13]. A 4 m course was used.

Comprehensive geriatric assessment
During Visit 1, comprehensive geriatric assessment (CGA) was performed by a single geriatrician (CW). This included Mini-Nutritional Assessment Short Form (MNA-SF) [14], Addenbrooke's Cognitive Examination-III (ACE-III) [15], evaluation of basic [16,17] and instrumental activities of daily living (ADLs) [18], Geriatric Depression Scale Five-Point (GDS-5) [19,20], and questioning on falls and admissions within the last year. Comorbidities and symptomatic burden were used to calculate the Geriatric Index of Comorbidity (GI-C) [21]. Frailty was defined by Fried frailty phenotype [22,23] and Clinical Frailty Scale [24]. Delirium screening with 4AT was performed at each visit [25]. Variables measured were considered potential effectors of acute sarcopenia [5]. These were included to assess feasibility and acceptability of measurement, and document individual differences ( Fig. 1).

Sarcopenia definition
This study was completed prior to EWGSOP2 dissemination; sarcopenia was defined using initial EWGSOP definition by low BATT (muscle mass) and either low handgrip strength or gait speed [1]. Cut-off values were taken two SDs below young healthy reference population means. Cut-off values were 3.86 cm (women) and 5.44 cm (men) for BATT [10], 20 kg (women) and 30 kg (men) for handgrip strength [13], and 0.8 m/s (both genders) for gait speed [13]. Low BATT alone was defined as pre sarcopenia; severe sarcopenia was defined as all three criteria below cut-offs [1].

Enzyme-linked immunosorbent assays (ELISAs)
Blood samples were taken at Visit 1 with serum prepared within 30-60 min of collection and stored at − 80 °C. Cortisol, dehydroepiandrosterone sulfate (DHEA-S), and high sensitivity C-Reactive Protein (hsCRP) (IBL International; Hamburg, Germany), and vitamin D (Immundiagnostik AG 25(OH)-vitamin D direct day ELISA; Bensheim, Germany) were measured by ELISA using commercial kits. Optical density was measured immediately using a photometer at 450 nm (Sci-Tek Instuments Ltd Synergy HT).

Data management and analysis
Data were entered into Microsoft Excel 2010, converted to IBM SPSS Statistics 22, and visually assessed for normality. Significance of differences between serial BATT, BATT-SCR, echogenicity, and handgrip strength were calculated using Friedman and post hoc Wilcoxon signed rank tests. Significance of difference in gait speed was calculated using Wilcoxon signed rank testing. Correlation between baseline immune-endocrine markers and changes in BATT, handgrip strength, and gait speed were assessed using Kendall's correlation. Difference in pre sarcopenia, sarcopenia, and severe sarcopenia rates between Visit 1 and Visit 3 were analysed using Chi squared test.
Measurements not obtained at any visit were analysed as missing values.

Ethical considerations
Ethical approval was obtained from Wales REC 6 (17/ WA/0117). Written informed consent was obtained from participants. Specific optional consent was obtained to consent to remain included in the event they lost capacity to consent during the study. If this were to occur, separate consultee declaration was planned to be obtained from nominated consultees.

Results
17 patients were approached to participate in this study; eight were recruited. One participant withdrew following Visit 1; data were not analysed. One participant declined handgrip strength testing at Visit 2, and one participant declined Visit 3 assessment. Visit 3 was conducted on median day 7 (IQR 5-8). Additional file 1 shows recruitment and follow-up flowchart for this study.

Changes in muscle mass and function
BATT and BATT-SCR differed between visits

Acute sarcopenia
Prevalence of sarcopenia at baseline was 28.6%. As assessments were not performed on one participant at Visit 3, incidence of acute sarcopenia at Visit 3 was 75%. Overall prevalence of sarcopenia at Visit 3 was 83.3%. One participant, who had pre sarcopenia at Visit 1, developed severe sarcopenia. Changes were not statistically significant (p = 0.558). Results are shown in Fig. 1d.

Feasibility and acceptability of measurements
It was possible to perform all muscle assessments in all participants at Visit 1. BATT was measured at each visit with participants, although one participant declined all assessments at Visit 3. One participant declined handgrip strength assessment at Visit 2. SPPB was measured at baseline; only gait speed was measured at Visit 3, due to potential harm from abdominal strain in performing chair stands. Nutritional, mood, functional, and comorbidity assessment were performed in all participants at Visit 1. Cognitive screening was initially performed, but following withdrawal of one participant due to burden of assessment, a decision was made to remove formal cognitive testing. Participants expressed they found all other assessments acceptable. No participant exhibited loss of capacity, but all participants consented to ongoing inclusion in the event they should have lost capacity during the study.

Discussion
Prevalence of sarcopenia at Visit 1 was similar to previous studies in hospitalised older patients [26][27][28][29]. Reduction in BATT and gait speed, and increased prevalence of sarcopenia at Visit 3 compared to Visit 1 was demonstrated. Muscle strength change showed greater variability between participants. The next stage of our research will characterise a larger cohort, including elective and emergency admissions, and include three-month followup. We have shown that measurement of these variables is feasible in an acute care setting. Due to perceived burden of ACE-III, and following further public involvement, this will not be measured in our main study. This should enable time to focus on nutritional and physical function assessment, and comparison measurements of muscle mass with BIA. BATT increased at Visit 2 in all but one participant; BATT-SCR reduced across visits for all participants. This suggests increased BATT was associated with increased SC thickness. We consider increased SC thickness more likely represents oedema than increase in fat tissue. Similar findings have been demonstrated in critical care patients following cardiothoracic surgery, with quadriceps muscle thickness correlating with fluid balance [30]. However, changes in fluid balance alone may be insufficient to explain this. Fluid shifts may have been exacerbated by systemic inflammation and increased vascular permeability postoperatively [31]. Considering increases in BATT within 24 h of surgery may represent heightened systemic inflammation, this should not be disregarded as a benign process. If acute increase in BATT relates to increased inflammation, this could herald increased likelihood of developing acute sarcopenia. Our planned larger study will assess if rate of change to Visit 2 predicts rate of change to Visit 3.
Baseline hsCRP correlated with change in BATT to Visit 3, suggesting older adults with lower baseline inflammation were at greater risk of acute reductions in muscle size. This is converse to associations demonstrated in chronic sarcopenia; higher levels of hsCRP and proinflammatory cytokines have been demonstrated in older adults with sarcopenia [32] although, a positive association between hsCRP and lean appendicular mass was demonstrated in a large cohort study [33]. Baseline DHEA-S levels correlated with change in gait speed; agerelated reduced DHEA-S has been shown to correlate with reduced physical performance [34]. We will further evaluate this in our larger study; a proven association may suggest exogenous DHEA-S could be used to protect older adults from acute declines in physical performance.
Our main study will fully characterise acute muscle changes, and acute sarcopenia in elective and emergency patients. This will include CGA, particularly nutritional and physical function assessment, and measurement of biomarkers of immune-endocrine function, as guided by our pilot results. Follow-up will be conducted to assess if acute changes in muscle mass and function are predictive of change in physical function at 3 months.

Limitations
Firstly, we recognise that, although tests of statistical significance were performed, this pilot study was not powered to identify changes in muscle mass or function. We fully acknowledge that no definite conclusions can be made from this study alone. However, we consider these results may be of value in future research and, potentially, used in a systematic review. Our results have reaffirmed the need for our larger study, and assisted with power calculations.
An important characteristic that differed from expected population within this study was gender. Percentage of elective colorectal surgery procedures performed on females has been reported as 41.7-56.2% [35][36][37]. This is important as muscle echotexture, muscle: SC tissue