Ultrasonography to measure quadriceps muscle in critically ill patients: A literature review of reported methodologies

Abstract

Muscle wasting in the intensive care unit (ICU) is common and may impair functional recovery. Ultrasonography (US) presents a modern solution to quantify skeletal muscle size and monitor muscle wasting. However, no standardised methodology for the conduct of ultrasound-derived quadriceps muscle layer thickness or cross-sectional area in this population exists. The aim of this study was to compare methodologies reported for the measurement of quadriceps muscle layer thickness (MLT) and cross-sectional area (CSA) using US in critically ill patients. Databases PubMed, Ovid, Embase, and CINAHL were searched for original research publications that reported US-derived quadriceps MLT and/or CSA conducted in critically ill adult patients. Data were extracted from eligible studies on parameters relating to US measurement including anatomical location, patient positioning, operator technique and image analysis. It was identified that there was a clear lack of reported detail and substantial differences in the reported methodology used for all parameters. A standardised protocol and minimum reporting standards for US-derived measurement of quadriceps muscle size in ICU is required to allow for consistent measurement techniques and hence interpretation of results.

Keywords

Critical illness ultrasound quadriceps muscle size muscle thickness cross-sectional area muscle wasting

Introduction

Critically ill patients can experience extensive muscle wasting that occurs rapidly from intensive care unit (ICU) admission.¹ Muscle wasting may contribute to the condition known as ICU-acquired weakness (ICU-AW) with consequent effects on strength and functional outcomes. Quantifying muscle size may therefore identify patients at risk of ICU-AW and poor functional recovery, and help guide interventions to combat atrophy and weakness.²

The most direct methods to measure musculature are computed tomography (CT) and magnetic resonance imagining (MRI), however their use in critical illness is complicated by a lack of deployability, repeated radiation exposure and incompatible monitoring equipment.³ In addition, CT scans of the third lumbar region to measure muscularity have been reported more recently in the literature,⁴ however this remains a research appropriate rather than clinically relevant technique, due to radiation exposure preventing serial measurements. Accordingly, ultrasonography (US) has gained momentum as a non-invasive, bedside technique for the quantification of muscle size in these patient populations.⁵

One of the barriers of using US to measure musculature is that differences in technique methodology can affect the results obtained. In order to maintain reliability of musculature measurements, a defined, standardised approach is required. Consistency in technique to prevent measurement error is required including type of probe, probe positioning, patient positioning, landmarking, choice of equipment, technical settings such as gain, pressure application and distortion minimisation. To our knowledge, there is currently no internationally accepted validated guideline or framework for the measurement of musculature in critical illness using ultrasound. While there are tutorial guides available in other populations,^6,7 it should be recognised that these are not validated protocols and there is a clear lack of consensus in reporting US techniques and study methodologies. Furthermore, critically ill patients are a unique population with factors, such as fluid shifts, that influence the translation of evidence from other populations.⁸

The quadriceps is the most extensively studied weight-bearing muscle group, with strong correlations to muscle mass and strength in a number of patient groups.⁹ Muscle size of the quadriceps can be measured by either cross-sectional area (CSA) and/or muscle layer thickness (MLT). Measurement of CSA involves calculating the area of a target structure from a frozen ultrasound image.¹⁰ MLT utilises on-screen callipers or programs to calculate the thickness of structure(s) of interest: either total quadriceps thickness or thicknesses of specific muscle groups. Within the literature, there appears to be no consensus on which method is more representative of true muscle size. Although Puthucheary et al. suggest that MLT underestimates muscle wasting and strength,¹¹ studies rarely explore benefit of one over the other.

As standardisation of US imaging and reporting is essential to the future validity and application of US musculature assessment, the aim of this systematic review is to compare the similarities and differences in published methodologies for US-derived measurement of quadriceps musculature in critically ill patients.

Methods

Electronic databases were searched on 31 January 2019 by a single reviewer (LW) using a systematic and reproducible approach as follows. Potential studies were identified using pre-existing literature databases and predefined search criteria (Table 1). Studies were then screened to identify eligible studies for review (LW). This involved the removal of duplicate publications and assessment of eligibility against the inclusion and exclusion criteria from title and abstract.

Table 1.

Literature search criteria.

Database name	Search fields	Search terms
PubMed, Ovid, Embase and CINAHL (EBSCOhost)	Text Word	1. Ultrasound OR Ultrasonography2. Muscle OR quadriceps OR femur OR thigh3. Critically ill, OR Critical care OR Critical illness OR ICU OR Intensive Care4. No. 1 and No. 2 and No. 3

ICU: intensive care unit

Study inclusion criteria

Studies were eligible for review if they:

were a published peer-reviewed original research article with a quantitative study design;

studied critically ill adult participants admitted to an ICU;

included ultrasound measurement of the quadriceps muscle for MLT and/or CSA; and

were written in the English language

No restrictions were placed on studies with or without a control arm or on year of publication.

Study exclusion criteria

Studies that solely reported other ultrasound-derived measures such as echogenicity (muscle quality assessment) or pennation angle (force contractility), or measurement of muscle groups other than the quadriceps were excluded.

Where further clarification was required to assess relevance, the full-text was obtained (LW). Data extraction was then carried out (LW and LC) using predefined characteristics (Table 2). Any discrepancies in data extraction were settled by a third data extraction (MS). Figure 1 shows the review process.

Table 2.

Data extraction parameters.

Reported parameters	Method of reporting
Critically ill subpopulation	Inclusion and exclusion criteria
Time of assessment	Time of US performance
What was analysed	Muscle layer thickness or cross-sectional analysis
Ultrasound type	Brand/mode
Probe/transducer used	MHz and probe type
Patient positioning	Positioning of body, limbs, elevation, extension
Anatomical location of measurement	Landmarking for location of scanning
Plane of measurement	Plane that the musculature was scanned in
Probe positioning	The positioning of the probe when held against the musculature
Number of measurements	Number of images taken per muscle, scan and or per patient
Averaging of measurements	Averaging of multiple measurements within patients
Exclusion of images	Methods and reasoning for exclusion of captured images
Pressure of probe	No pressure, minimal or maximal pressure
Distortion elimination^a	Reference made to reducing or eliminating distortion and methods as to how distortion was eliminated
Depth	Depth of scanning
Other technical settings	Gain, focus, mechanical index, acquisition settings
Landmark marking	Methods of marking location for scanning
Number of operator(s)^b	Number of US operators
Number of assessor(s)^c	Number of image operators
Same operator(s) as assessor(s)	Same US operator also assessing musculature
Operator or assessor blinding	Operators of US measurement blinded to each other, intervention, or to assessor measurement of muscle size as relevant
Measurement side	Side that US measurements were conducted
Image scoring techniques	Techniques to score images for outcomes
Inter- or intra-rater reliability	Inclusion of inter- and/or intra-rater reliability of US measurement

^aAny method that makes direct reference to ‘elimination of distortion’ with an associated reduction method.

^bIndividual(s) responsible for image acquisition only.

^cIndividual(s) not involved in image acquisition and only in MLT/CSA measures.

US: ultrasonography; MLT: muscle layer thickness; CSA: cross-sectional area.

Figure 1.

Literature search results following initial search, screening and number of studies included in analysis. ICU: intensive care unit; MLT: muscle layer thickness; CSA: cross-sectional area.

Results

There were 4028 papers identified following the literature search. Following the screening process 36 papers were included in the final data extraction and analysis (Figure 1, Table 3).

Table 3.

Ultrasound protocol data extraction for measuring the quadriceps musculature.

Author and year	Study design	Timepoint/s of assessment (day of admission unless otherwise stated)	Measure	Pressure applied	Muscle group(s)	Anatomical landmark	Distance
Muscle thickness
Baldwin, 2014	Prospective cross-sectional	Two measures taken < 1 day apart. Timepoint not stated	MLT	NS	Mid-thigh	NS	NS
Cartwright, 2013	Prospective pilot	Days 0 (within 80 hrs), 3, 7, 14 and months 2, 4, 6 after H/D	MLT	Minimal	RF	From superior portion of patella	15 cm proximal
Chapple, 2017	Prospective observational	Within 7 days, weekly until H/D	MLT	Maximal	Quads	ASIS and upper pole of patella	Lower one-third and upper two-thirds and midpoint
Fetterplace, 2018	Prospective randomised	Before randomisation, ICU D/C or day 15	MLT	NS	Quads	ASIS to upper patellar border	Midpoint and two-thirds
Fischer, 2016	Prospective, randomised, controlled, single-blind	Cohort A and B: Postop (Cohort A additional assessment pre-surgery), every other day until ICU D/C and at H/D	MLT	Minimal	Quads	1. Lateral point: superior border of greater trochanter and lateral knee joint space2. Ventral point: on a line from the ASIS to the superior border of the patellar base, on the same level as the lateral point3. Medial point: On a line from the pubis to the medial knee joint space, on the same level as the lateral point	1. Midpoint2. NS 3. NS
Gruther, 2008	Two-fold setting: prospective longitudinal and cross-sectional single-blind	Group A: days 2, 28Group B: Once-off	MLT	NS	RF, VI	ASIS to upper pole of the patella	Lower and upper two-thirds, midpoint
Gruther, 2010	Randomised, controlled, double-blind, pilot	Baseline, 4 weeks later	MLT	NS	RF, VI	ASIS to upper pole of the patella	Lower and upper two-thirds, midpoint
Hadda, 2017	Cross-sectional	Not stated	MLT	Minimal	Quads	Tip of greater trochanter and lateral joint line of the knee	Midpoint
Hadda 2018	Prospective observational	Days 1, 3, 5, 7, 10, 14, then weekly until D/C or death	MLT	NS	Mid-thigh	Greater trochanter to lateral joint line of knee	Midpoint
Katari, 2018	Observational	Day 1 of ICU adm, days 3 and 7	MLT	No pressure	RF, VI	AIIS to superior border of patella	NS
Kelmenson, 2018	Prospective cohort	Weekly until D/C, death, developed critical illness polyneuromyopathy, or 4 weeks of measures	MLT	Minimal	Mid-thigh	ASIS to superior border of patella	Midpoint
Paris, 2017	Prospective multicentre observational	Within 72 h of CT scan	MLT	Maximal	RF, VI	ASIS and upper pole of patella	Lower and upper two-thirds, midpoint
Reid, 2004	Prospective	Every 1–3 days	MLT	NS	Thigh	Greater trochanter to lateral joint line of knee	Midpoint
Sabitino, 2017	Cross-sectional observational	Start and end of each RRT or dialysis session	MLT	Minimal	RF, VI	NS	NS
Sarwal, 2015	Cross-sectional observational	Random LOS	MLT	NS	RF, VI	ASIS to superior patellar border	Two-thirds
Silva, 2017	Observational	Days 1 (within 24 h of MV), 7, 14	MLT	Minimal	RF	Iliac spine to superior border of patella	Midpoint
Turton, 2016	Observational	Days 1 (within 24 h of MV), 5, 10	MLT	Minimal	VL	Lateral condyle of femur	10 cm proximal
Witteveen, 2017	Cross-sectional	Upon ICU awakening	MLT	NS	RF	ASIS to superior border of patella	Midpoint
			CSA
Annetta, 2017	Prospective observational	Days 0 (within 24 h of trauma) 5, 10, 15, 20	CSA	NS	RF	ASIS to superior border of patella	Three-fifths from ASIS
Berger, 2018	Prospective observational	Days 4, 9, 10, 15	CSA	NS	RF	NS
Bloch, 2013	Prospective longitudinal observational	Pre-op, day 7 or H/D	CSA	NS	RF	ASIS to superior border of patella	Three-fifths from ASIS (66% or 80% if unable to visualise)
Borges, 2018	Prospective cohort	Days 2, 4, 6, ICU D/C, H/D	CSA	NS	RF	Superior iliac crest and upper edge of patella	Three-fifths
Connolly, 2018	Prospective	2×, separated by 7 days	CSA	Minimal	RF	ASIS to superior patellar border	Two-thirds from ASIS
Gerovasili, 2009	Prospective	Day of randomisation (second day of adm), day 7 or 8	CSD	Minimal	RF, VI	ASIS to patella	Midpoint
Haaf, 2017	Observational, pilot	Within 72 h, weekly for 4 weeks	CSA	NS	RF, VI	ASIS and proximal border of patella	Midpoint
Mueller, 2016	Prospective observational	On enrolment (within 72 h of ICU admission?)	CSA	Minimal	RF	ASIS to superior border of patella	Three-fifths from ASIS
Pardo, 2018	Prospective observational	Days 1, 2, 5, 7, 21	CSA	Maximal	RF	ASIS to upper border of patella	Midpoint and two-thirds
Puthucheary, 2013	Prospective	Days 1 of hospital admission, 10	CSA	No pressure	RF	NS	NS
Puthucheary, 2015	Prospective longitudinal two centre observational	Days 1, 3, 7, 10	CSA	NS	RF	NS	NS
Twose 2018	Prospective observational	≤ 24 hrs of MV, day 3, 5, 7, 10, or until no longer MV	CSA	Minimal	RF	ASIS to superior patellar border	Two-thirds
			MLT and CSA
Ferrie, 2016	Double-blinded randomised controlled	On enrolment, days 3, 7	CSAMLT	Minimal	RF Quads	NSSuperior greater trochanter to superior lateral border of tibia head	NSMidpoint and two-thirds
Galindo, 2018	Prospective observational	≤ 48 hrs of ICU adm	CSAMLT	Maximal	RF, VIVL, VI	AIIS to proximal border of patellaMoved laterally 6–12 cm from RF–VI measure	Midpoint and two-thirdsMidpoint
Hayes, 2018	Prospective cohort	Day 1, 10, 20	CSAMLT	NS	RF, VI	Medial ASIS to superior patella border and b/w femoral epicondyles	Midpoint and two-thirds
Mukhopadhyay, 2018	Prospective observational cohort	Day 1, 2, 7, 10 and ICU adm	CSAMLT	NS	RF	AIIS to the superior patella border	Two-thirds
Palakshappa, 2018	Prospective cohort	Within 48 h of ICU adm and within 48 h of d7 after ICU adm	CSAMLT	Minimal	RFVI	ASIS to proximal patella	Two-thirds
Parry, 2015	Single-centre prospective observational	Baseline (early as possible), days 3, 5, 7, 10, awakening, and ICU D/C	MLTCSA	No pressure	RF, VI	ASIS to superior patella border	Two-thirds and 2 cm laterally

NS: not stated; MLT: muscle layer thickness; CSA: cross-sectional area; RF: rectus femoris; Quads: quadriceps; VL: vastus lateralis; VI: vastus intermedius; ASIS: anterior superior iliac spine; LOS: length of stay; RRT: renal replacement therapy; H/D: hospital discharge; Pre-op: preoperative; adm: admission; Postop: postoperative; ICU: intensive care unit; D/C: discharge; AIIS: anterior inferior iliac spine; MV: mechanical ventilation; CT: computed tomography; LOS: length of stay.

Patient population and study design

Inclusion/exclusion criteria

Four papers did not exclude patients with pre-existing conditions specifically noted to have a detrimental effect on peripheral muscle size.^5,12–14

Timepoint of assessments

All but seven^8,14–19 authors reported serial US measurements. Of these, only six explicitly reported being conducted after ICU discharge^5,20–24 (Table 3). The large majority of studies were observational with only six having a randomised study design.^{12,13,22,25–27}

Patient related

Positioning, contraction/relaxation and rotation

The majority of studies reported supine positioning, with eight not referencing any patient positioning.^{17,21,22,27–31} Mueller et al. and Turton et al. referenced the head of bed at a 30° and 45° elevation, respectively,^16,32 while one paper reported a 30°–45° head of bed elevation,¹⁹ and another referenced semi-supine positioning.³³ The term ‘limb extension’ was reported by 21 studies.^{5,8,12–16,19,23–26,33–41} One paper referenced ‘legs flat’ and was not included as referencing extension.²⁰ Relaxation of limbs was reported by 11 authors,^{5,13,14,16,20,26,35,36,38,39,41} with one using a pillow under the knees for muscular relaxation.²² Four studies reported foot positioning; toes facing the roof.^8,19,36 Neutral rotation/positioning of the lower limbs was reported by five authors^{12,15,23,33,37} and external rotation of the lower limb by one.³²

Ultrasound procedures

MLT was reported by 18 articles, 12 reported CSA, and six reported both CSA and MLT (Table 3). For MLT four papers measured the rectus femoris (RF), one the vastus lateralis (VL), one the vastus intermedius (VI), one measured VL and VI, and five measured total quadriceps MLT. Four studies reported the ‘mid-thigh’ or ‘thigh’ with no further detail of anatomical site of measurement (Table 3). For CSA 13 papers measured RF and five measured RF and VI. Landmarking of anatomical locations and thickness measures were reported by the majority of authors, however the location descriptors were variable (Table 3). The side that image acquisition was carried out was generally bilateral^{5,13,14,21–23,26,31,34–36,38,39,41,42} or on the right.^{8,12,15–20,32,33,43,44} No studies analysed solely the left side aside from Hayes et al. (2018)³⁷ taking measures from the non-cannulated leg with no further detail of the side used. The majority of studies did not report how landmarking of anatomical sites was performed. Out of the 29 studies that conducted serial measures, only half of these reported the use of a marker to ensure accuracy for consequent acquisitions.

Measurement procedures

The amount of pressure applied to the US probe for image acquisition was largely reported as minimal (15/36 articles) (Table 2). Elimination of distortion was referred to in six studies; five stated the use of excess gel,^{14,25,28,34,36} while the remaining study referenced gentle contraction, relaxation techniques and passive mobilisation to help visualise structures.³³ US probe positioning was largely uniform with perpendicular positioning being the most predominant (Table 3). The plane of acquisition was not referenced in 19 studies.

US technical settings

The depth to which the US image was taken was reported by 12 authors. Of these, three reported a specific depth—5,¹⁵ 1.9–6.0⁴¹ and 4.9 cm³⁷—with the others referring to either a standard/constant setting (e.g. always at 6 cm depth)^21,34 or alteration to visualise the whole structure(s) of interest.^{14,17,28,32,33,38,39} Other settings (gain, focus and contrast) were made reference to in ten studies^{13,15,17,21,23,28,33,34,37,38} with only two providing more information than standardisation of or constant settings.^21,33

Type of US transducer used

The type of US transducer used was reported by 30 studies (Table 4). Sound wave frequency (MHz) and the array of the transducer footprint used varied. Linear array probes were reported in half of the studies, with three reporting a curvilinear probe and one the use of a sector array when the thigh cross section could not be visualised completely.

Table 4.

Ultrasound protocol data extraction for ultrasound transducer procedures and equipment.

Author and year	Plane	Probe positioning	US Mode	US type	Transducer frequency (MHz)	Footprint (array)
Annetta, 2017	NS	Perpendicular to long axis of muscle	NS	Esaote MyLab	5–7.5	NS
Baldwin, 2014	NS	NS	NS	NS	NS	NS
Berger 2018	NS	NS	NS	NS	NS	NS
Bloch, 2013	Cross-sectional	Perpendicular to leg axis	B-mode	Portable Logiq E, GE	1012L-RS	NS
Borges, 2018	NS	Perpendicular along thigh shaft	NS	GE Vivid 7	7.5	Linear
Cartwright, 2013	Cross-sectional, sagittal	Perpendicular to muscle	NS	Esaote Biosound MyLab 25	18	Linear
Chapple, 2017	NS	Perpendicular to skin	B-mode	SonoSite X-Porte	13–6	NS
Connolly, 2018	Oblique cross-section	NS	B-mode	NS	2–6	Curved
Ferrie, 2016	Perpendicular to long axis	Perpendicular to limb surface	2D	Sonosite M-Turbo	MF	Linear sector (large patient limb size)
Fetterplace, 2018	NS	Perpendicular to skin	NS	Sonosite S-ICU	13–6	NS
Fischer, 2016	Transverse, sagittal plane	NS	B-mode 2D	GE Vivid i	6	NS
Galindo, 2018	NS	Perpendicular to skin	B-mode	NS	13	NS
Gerovasili, 2009	Ventral to transverse plane	Perpendicular to bone surface	NS	GE Vivid 7 model	7.5	Linear
Gruther, 2008	NS	NS	NS	Portable ATL (HDI-1000)	NS	Linear
Gruther, 2010	NS	NS	NS	Portable ATL (HDI-1000)	NS	Linear
Haaf, 2017	Transverse	Perpendicular	NS	Sonosite M-Turbo	13	NS
Hadda, 2017	Anterior	Perpendicular to skin	B-mode	Siemens ACUSON X300	5–13	Linear
Hadda, 2018	NS	Perpendicular to skin	B-mode	Siemens ACUSON X300	5–13	Linear
Hayes, 2018	NS	Perpendicular to skin	NS	Sonosite Edge Portable	6–15	Linear
Katari, 2018	Transverse	Perpendicular to long axis of muscle	B-mode	Sonosite M-turbo	5–10	Linear
Kelmenson, 2018	Transverse	Perpendicular to underlying bone	NS	Philips Sparq	NS	Linear
Mueller, 2016	Transverse	Perpendicular along superior aspect of thigh	B-mode	Philips	NS	NS
Mukhopadhyay, 2018	NS	Perpendicular to long axis of thigh	B-mode	NS	NS	Linear
Palakshappa, 2018	Transverse	Perpendicular to long axis of thigh	NS	Sonosite Edge Portable	5–10 MLT 2–5 CSA	LinearCurved
Pardo, 2018	NS	Perpendicular to long axis of thigh	NS	GE Vivid i	12	Linear
Paris, 2016	NS	Perpendicular	B-mode	NS	MF	Linear
Parry, 2015	NS	Perpendicular	2D	Voluson e BT09	8.5	Linear
Puthucheary, 2013	NS	NS	B-mode	NS	NS	NS
Puthucheary, 2015	NS	NS	B-mode	NS	NS	NS
Reid, 2004	Cross-sectional	NS	NS	ALOKA Echo	5	NS
Sabitino, 2017	NS	Perpendicular to long axis	B-mode	NS	7.5	Linear
Sarwal, 2015	Cross-sectional	Perpendicular to long axis of thigh	NS	Not stated	8.5	Linear
Silva et al., 2017	Transversal	Perpendicular to tissue interface	B-mode	Sonosite M-turbo	7.5	NS
Turton et al., 2016	Plane with muscle	Parallel to ground	B-mode	NS	NS	NS
Twose, 2018	NS	Perpendicular to long axis of thigh	NS	Sonosite X-Porte	5–2	Curved
Witteveen et al., 2017	Transverse	Femur located in middle with VI and RF visible	B-mode	Esaote MyLab Twice	4–13	NS

MF: multi frequency; NS: not specified: Quads: quadriceps; RF: rectus femoris; VI: vastus intermedius; US: ultrasonography.

Operators

The operator was defined as an individual(s) conducting image acquisition only at the bedside, while an assessor solely conducted image analysis. For those studies that included more than one operator (n = 28), three articles clearly stated that the operator was not blinded^16,22,45 and eight that operators were blinded.^{8,12,15,25,26,35,36,42} The majority of studies only included one operator (19/37; 51%) with six having two, ^{8,17,24,25,36,42} one having three³² and the remaining not reporting this parameter.^{14,15,18,20,22,27–30,40} Experienced^{15,21,23,33,39} or trained^{5,8,12,13,16,22,30,32,34,36–38,41,42} operators were referenced in 19 papers.

Image analysis

Number of images at each anatomical site

Thirteen authors did not state the number of US images obtained^{8,15,16,18,21,25,27,29–31,38,42,43}. Four studies acquired measures in duplicate.^13,14,34,36 Triplicate measures were recorded by 13 authors^{8,12,17,20,22–24,28,33,37,40,41,45} while three took measures in triplicate only if there was a difference of >10%^5,39 or 0.1 cm.¹⁹ The remaining authors reported taking one,²⁶ six,³² or ‘the best three images’.³⁰

Measurement tools

There was great variation in the measurement tool used to analyse the images. Nine articles used inbuilt callipers on the US machine,^{8,13,14,16,19,24,34,41,45} nine used Image-J,^{15,20,21,23,30,32,33,37,40} one used Photoshop Elements version 12²⁸, one used EchoPAC (General Electric Healthcare)²², and the remaining 16 did not state the method used.

Scoring of images

An assessment of muscle wasting using a scoring method in order to allow for interpretation of muscle size was only reported by two authors.^17,20 Reasons for exclusion of images included excessive oedema not allowing definition of the RF,^16,20,25,37 images without standardised gain settings²⁸ and removal of images with the highest and lowest width.³²

Assessors

In seven studies the same person acquired the US image (operator) and measured MLT/CSA (assessor).^{13,16,19,33,39,42,45} In five studies, the operator was a different person to the assessor,^{12,15,22,26,35} and all assessors were blinded to the US image. Of the six interventional studies, four studies reported that the person conducting the ultrasound measurement was blinded to the intervention.^12,22,25,35 Of the 30 studies that included serial US measures, only four reported that the investigator taking the US image was blinded to previous measures^12,22,30,37 and two explicitly reported that no blinding to previous measures occurred.^25,32

Inter- and intra-rater reliability

Inter-rater reliability was reported in 11 studies^{8,14,16,18,26,28,30,36,37,42,45} and intra-rater reliability was reported in seven studies.^{8,14,15,22,24,36,42} The most common method of reporting reliability was using intra-class coefficients.

Discussion

This systematic review identified 36 studies that analysed quadriceps muscle thickness or cross-sectional area in critically ill patients using US. Across these studies, it was observed that there was a clear lack of agreement in the methodological approach used, and the reporting of such, in regards to the US technique.

This review identified that the most common method of reporting musculature using US in the ICU literature is MLT, as opposed to CSA, although the authors generally provided little justification for the chosen methodology. CSA measurements have been shown to be a more reliable indicator of muscle wasting and strength in 19 patients up to day 10 of ICU admission.¹¹ However, MLT may have greater translation into a clinical setting when compared to CSA, with greater feasibility as results can be obtained quickly and in real time, as reported by Paris et al.¹⁴ Further, MLT may be easier to identify than CSA in the setting of muscle wasting, where distinction between muscle groups to allow CSA measurements may not be feasible. The development of an ultrasound protocol for use in critical care requires consideration of both the feasibility of measurement, as well as quantification of the amount of muscle on admission, and degree of muscle loss during hospitalisation that has an impact on outcomes.

Of the included studies, while CSA was primarily of the RF muscle, there was great variation in the conduct of linear measures, with both individual muscle groups and total muscle layer being measured. Of even greater concern is the inconsistency in the reporting of anatomical sites of measurements. While the majority of studies measured between the anterior superior iliac spine (ASIS) and patella, this was not unanimous. Further, the distance at which the US measurement was taken between these two sites varied greatly. Measures were taken at one-third, two-thirds, one-half, three-fifths, 10 or 15 cm proximal to the patella, or at a combination of sites. With no consensus on the appropriate anatomical site to conduct the measurement either the most accessible,⁴⁶ or methods translated from non–critically ill populations, have been referenced by the included papers.^18,34 This discrepancy in anatomical site of measurement is similarly reflected in tutorial guides: the most commonly proposed method is to measure muscle at two sites—at two thirds and the midpoint between the ASIS and the top of the patella^6,46—yet the anterior inferior iliac spine has also been proposed.⁴⁷ Little justification behind the proposed sites is provided. It is likely that using more than one landmark provides a more global, and hence accurate, reflection of muscle size and subsequently a better representation of muscle atrophy.⁴⁸ A consensus on appropriate landmarking technique is required to ensure muscle atrophy is being captured optimally and that consequent studies are both relatable and comparable. This is particularly important as we move forward to demonstrate relationships between interventions and degree of muscle wasting.

Although the majority of studies reported applying minimal pressure on the muscle group measured, the question of the correct amount of pressure to apply during image acquisition remains unanswered, with alternating theories for each. Critically ill patients are susceptible to conditions that may complicate visualisation of muscle wasting such as oedema.⁴⁹ Correct pressure applied to a US probe may improve image quality and therefore muscle measurements in the presence of these complications.⁵⁰ It has been hypothesised that minimal pressure may be easier to standardise across operators and may be able to identify, and therefore quantify, oedema.⁶ Surprisingly though, other reports state that maximal and not minimal compression is able to better allow for standardisation⁵¹ and muscle quantification in the presence of oedema.⁴⁸ These factors have not yet been thoroughly investigated⁶ and consideration of the ability to maintain consistent pressure in serial measurements is required. This dilemma provides a potential application for technology to standardise pressure application in the setting of muscle wasting.

A total of 29 studies conducted serial measures. For studies that report change in muscle size over time it is vital that the landmarking technique used encompasses appropriate strategies to ensure serial measures are acquired from the same location and in the same manner, particularly with regard to patient positioning and the degree of pressure applied.

Of the 36 studies, 19 reported that the person conducting the ultrasound was experienced. This is a concern given taking accurate US measures are highly dependent on training and experience.¹⁵ Although some studies have demonstrated reliability between muscle size acquisitions, these have often been in healthy populations⁵² for which measurements are likely to be less challenging than for the critically ill, without barriers such as positioning, fluid retention and wounds.¹⁵ Inter-rater reliability needs to be further demonstrated in populations that experience muscle wasting, and take into consideration barriers to limb positioning which adds complexity to the visualisation of anatomical structures.

This review also identified inconsistencies in patient positioning for ultrasound measures. While the majority of studies placed the patient in a supine position, only two referenced head of bed elevation. Head of bed elevation is often employed for reasons such as high intracranial pressure and reduction of ventilator-associated pneumonia risk.⁵³ Hacker et al. demonstrated significant differences (P < 0.001) in RF CSA across four different hip flexion positions (0°, 20°, 30° and 60°).⁵⁴ Therefore, alterations in elevation can give significantly different readings. In addition, limb extension was often reported, although relaxation and rotation were reported by few studies. As muscle structure length and angle are altered by contraction^55,56 and rotation, alterations of these parameters will affect measures.

With the increasing use of US to measure muscle atrophy, determining what muscle size, or degree of change, relates to clinical outcomes and/or is considered to constitute muscle wasting in the critically ill population is essential. The development of ‘cut-offs’ that equate to clinically relevant muscle wasting, similar to that done by Mourtzakis et al. with CT in cancer patients,⁵⁷ is required in order for US of musculature to be appropriately interpreted and implemented into clinical practice. Only one study in this review provided discussion on the degree of muscle loss that may constitute clinically relevant muscle wasting: Bloch et al. classified those with greater than 9.24% muscle loss as having muscle wasting, determined by a Bland–Altman plot with the coefficient of variation.²⁰ There is currently no score to determine what constitutes muscle wasting in the critically ill population and further work is required in this area.

This systematic review has a number of strengths and limitations. The search criteria were limited to those articles that measured the quadriceps muscle, as justified earlier. However, it should be recognised that there have been recent reports on the association between protein delivery and change in ultrasound-derived forearm thickness¹² as well as other authors measuring muscle size in other muscle groups.⁴⁵ Further, the inclusion of other US-derived measurements such as echogenicity and pennation angle may provide additional information on muscularity in critical illness, however this was beyond the scope of this review. During data extraction, only data specifically reported by the authors was included in this review, and hence contacting the authors may have provided more completeness to the data obtained. Similarly, only methodology reported in the primary publication or online supplements was extracted. Some authors may have referenced techniques, however as it is difficult to ascertain which elements of the reference papers were included in the current publication, these details were not reported. The strengths of this review include the use of multiple databases and ranges of terminology to capture as many papers as possible, and a rigorous assessment of many aspects that may affect ultrasound measurement of musculature.

Conclusion

This review identified that there are substantial differences in the reported methodology used to measure CSA/MLT of the quadriceps muscle in critically ill patients. The reporting of US methodology was not comprehensive, often failing to provide a complete understanding of the specific methods used. Where methods were reported, the US methodology varied greatly between studies preventing consistency and comparison between studies. As a result, developing a standardised training ‘package’ in this area, that includes reliability testing and minimum reporting standards, would be of great benefit ensuring US users are not only adequately trained in measuring musculature but also to ensure that the way in which US methodology is reported allows for accurate data interpretation within and between studies and methodology replication.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Dirks

Hansen

van Assche

, et al. Neuromuscular electrical stimulation prevents muscle wasting in critically ill comatose patients. Clin Sci (Lond) 2015; 128: 357–365.

Batt

Mathur

Katzberg

HD.

Mechanism of ICU-acquired weakness: muscle contractility in critical illness.

Intensive Care Med 2017; 43: 584–586.

Kraniotis

Muscle mass changes in the critically ill patient: The role of imaging.

J Postgrad Med 2017; 63: 147–148.

Looijaard

Dekker

Stapel

, et al. Skeletal muscle quality as assessed by CT-derived skeletal muscle density is associated with 6-month mortality in mechanically ventilated critically ill patients. Crit Care 2016; 20: 386.

Chapple

Deane

Williams

, et al. Longitudinal changes in anthropometrics and impact on self-reported physical function after traumatic brain injury. Crit Care Resusc 2017; 19: 29–36.

Earthman

CP.

Body composition tools for assessment of adult malnutrition at the bedside: A tutorial on research considerations and clinical applications.

JPEN 2015; 39: 787–822.

Smith

Finnoff

JT.

Diagnostic and interventional musculoskeletal ultrasound: Part 1. Fundamentals.

PMR 2009; 1: 64–75.

Hadda

Khilnani

Kumar

, et al. Intra- and inter-observer reliability of quadriceps muscle thickness measured with bedside ultrasonography by critical care physicians. Indian J Crit Care Med 2017; 21: 448–452.

Mourtzakis

Parry

Connolly

, et al. Skeletal muscle ultrasound in critical care: A tool in need of translation. Ann Am Thorac Soc 2017; 14: 1495–1503.

10.

Seymour

Ward

Sidhu

, et al. Ultrasound measurement of rectus femoris cross-sectional area and the relationship with quadriceps strength in COPD. Thorax 2009; 64: 418–423.

11.

Puthucheary

McNelly

Rawal

, et al. Rectus femoris cross-sectional area and muscle layer thickness: Comparative markers of muscle wasting and weakness. Am J Respir Crit Care Med 2017; 195: 136–138.

12.

Ferrie

Allman-Farinelli

Daley

, et al. Protein requirements in the critically ill: A randomized controlled trial using parenteral nutrition. JPEN 2016; 40: 795–805.

13.

Fetterplace

Deane

Tierney

, et al. Targeted full energy and protein delivery in critically ill patients: A pilot randomized controlled trial (FEED Trial). JPEN 2018; 42: 1252–1262.

14.

Paris

Mourtzakis

Day

, et al. Validation of bedside ultrasound of muscle layer thickness of the quadriceps in the critically ill patient (VALIDUM Study). JPEN 2017; 41: 171–180.

15.

Sarwal

Parry

Berry

, et al. Interobserver reliability of quantitative muscle sonographic analysis in the critically ill population. J Ultrasound Med 2015; 34: 1191–1200.

16.

Mueller

Murthy

Tainter

, et al. Can sarcopenia quantified by ultrasound of the rectus femoris muscle predict adverse outcome of surgical intensive care unit patients as well as frailty? A prospective, observational cohort study. Ann Surg 2016; 264: 1116–1124.

17.

Witteveen

Sommers

Wieske

, et al. Diagnostic accuracy of quantitative neuromuscular ultrasound for the diagnosis of intensive care unit-acquired weakness: a cross-sectional observational study. Ann Intensive Care 2017; 7: 40.

18.

Baldwin

Bersten

AD.

Alterations in respiratory and limb muscle strength and size in patients with sepsis who are mechanically ventilated.

Phys Ther 2014; 94: 68–82.

19.

Galindo Martin

Ubeda Zelaya

RDC

Monares Zepeda

, et al. ROUNDS Studies: Relation of OUtcomes with Nutrition Despite Severity-Round One: Ultrasound muscle measurements in critically ill adult patients. J Nutr Metabol 2018; 2018: 7142325.

20.

Bloch

Lee

Wort

, et al. Sustained elevation of circulating growth and differentiation factor-15 and a dynamic imbalance in mediators of muscle homeostasis are associated with the development of acute muscle wasting following cardiac surgery. Crit Care Med 2013; 41: 982–989.

21.

Cartwright

Kwayisi

Griffin

, et al. Quantitative neuromuscular ultrasound in the intensive care unit. Muscle Nerve 2013; 47: 255–259.

22.

Fischer

Spiegl

Altmann

, et al. Muscle mass, strength and functional outcomes in critically ill patients after cardiothoracic surgery: does neuromuscular electrical stimulation help? The Catastim 2 randomized controlled trial. Crit Care 2016; 20: 30.

23.

Parry

El-Ansary

Cartwright

, et al. Ultrasonography in the intensive care setting can be used to detect changes in the quality and quantity of muscle and is related to muscle strength and function. J Crit Care 2015; 30: 1151.e1159-1114.

24.

Borges

Soriano

FG.

Association between muscle wasting and muscle strength in patients who developed severe sepsis and septic shock.

Shock 2019; 51: 312–320.

25.

Gerovasili

Stefanidis

Vitzilaios

, et al. Electrical muscle stimulation preserves the muscle mass of critically ill patients: A randomized study. Crit Care 2009; 13: R161.

26.

Gruther

Benesch

Zorn

, et al. Muscle wasting in intensive care patients: Ultrasound observation of the M. quadriceps femoris muscle layer. J Rehab Med 2008; 40: 185–189.

27.

Berger

Pantet

Jacquelin-Ravel

, et al. Supplemental parenteral nutrition improves immunity with unchanged carbohydrate and protein metabolism in critically ill patients: The SPN2 randomized tracer study. Clin Nutr. E-pub ahead of print 5 Nov 2018. DOI: 10.1016/j.clnu.2018.10.023.

28.

Puthucheary

Phadke

Rawal

, et al. Qualitative ultrasound in acute critical illness muscle wasting. Crit Care Med 2015; 43: 1603–1611.

29.

Puthucheary

Rawal

McPhail

, et al. Acute skeletal muscle wasting in critical illness. JAMA 2013; 310: 1591–1600.

30.

Silva

Maldaner

Vieira

, et al. Neuromuscular electrophysiological disorders and muscle atrophy in mechanically-ventilated traumatic brain injury patients: New insights from a prospective observational study. J Crit Care 2018; 44: 87–94.

31.

Mukhopadhyay

Tai

Remani

, et al. Nutritional risk assessment at admission can predict subsequent muscle loss in critically ill patients. European J Clin Nutr 2018; 72: 1187–1190.

32.

Turton

Hay

Taylor

, et al. Human limb skeletal muscle wasting and architectural remodeling during five to ten days intubation and ventilation in critical care: An observational study using ultrasound. BMC Anesthesiol 2016; 16: 119.

33.

Connolly

Maddocks

MacBean

, et al. Nonvolitional assessment of tibialis anterior force and architecture during critical illness. Muscle Nerve 2018; 57: 964–972.

34.

Annetta

Pittiruti

Silvestri

, et al. Ultrasound assessment of rectus femoris and anterior tibialis muscles in young trauma patients. Ann Intensive Care 2017; 7: 104.

35.

Gruther

Kainberger

Fialka-Moser

, et al. Effects of neuromuscular electrical stimulation on muscle layer thickness of knee extensor muscles in intensive care unit patients: A pilot study. J Rehab Med 2010; 42: 593–597.

36.

Sabatino

Regolisti

Bozzoli

, et al. Reliability of bedside ultrasound for measurement of quadriceps muscle thickness in critically ill patients with acute kidney injury. Clin Nutr 2017; 36: 1710–1715.

37.

Hayes

Holland

Pellegrino

, et al. Acute skeletal muscle wasting and relation to physical function in patients requiring extracorporeal membrane oxygenation (ECMO). J Crit Care 2018; 48: 1–8.

38.

Kelmenson

Quan

Moss

What is the diagnostic accuracy of single nerve conduction studies and muscle ultrasound to identify critical illness polyneuromyopathy: A prospective cohort study.

Crit Care 2018; 22: 342.

39.

Palakshappa

Reilly

Schweickert

, et al. Quantitative peripheral muscle ultrasound in sepsis: Muscle area superior to thickness. J Crit Care 2018; 47: 324–330.

40.

Twose

Jones

Wise

MP.

Effect of hypercapnia on respiratory and peripheral skeletal muscle loss during critical illness: A pilot study.

J Crit Care 2018; 45: 105–109.

41.

Ten Haaf

Hemmen

van de Meent

, et al. The magnitude and time course of muscle cross-section decrease in intensive care unit patients. Am J Physical Med Rehab 2017; 96: 634–638.

42.

Pardo

El Behi

Boizeau

, et al. Reliability of ultrasound measurements of quadriceps muscle thickness in critically ill patients. BMC Anesthesiol 2018; 18: 205.

43.

Katari

Srinivasan

Arvind

, et al. Point-of-Care ultrasound to evaluate thickness of rectus femoris, vastus intermedius muscle, and fat as an indicator of muscle and fat wasting in critically ill patients in a multidisciplinary intensive care unit. Indian J Crit Care Med 2018; 22: 781–788.

44.

Hadda

Kumar

Khilnani

, et al. Trends of loss of peripheral muscle thickness on ultrasonography and its relationship with outcomes among patients with sepsis. J Intensive Care 2018; 6: 81.

45.

Reid

Campbell

Little

RA.

Muscle wasting and energy balance in critical illness.

Clin Nutr 2004; 23: 273–280.

46.

Galindo Martín

Monares Zepeda

Lescas Méndez

OA.

Bedside ultrasound measurement of rectus femoris: A tutorial for the nutrition support clinician.

J Nutr Metabolism 2017; 2017: 1–5.

47.

Pasta

Nanni

Molini

, et al. Sonography of the quadriceps muscle: Examination technique, normal anatomy, and traumatic lesions. J Ultrasound 2010; 13: 76–84.

48.

Mourtzakis

Wischmeyer

Bedside ultrasound measurement of skeletal muscle.

Current Opin Clin Nutr Metab Care 2014; 17: 389–395.

49.

Pillen

van Alfen

Skeletal muscle ultrasound.

Neurol Res 2013; 33: 1016–1024.

50.

Boezaart

Ihnatsenka

Ultrasound: Basic understanding and learning the language.

Int J Shoulder Surg 2010; 4: 55.

51.

Pigula-Tresansky

Kapur

, et al. Muscle compression improves reliability of ultrasound echo intensity. Muscle Nerve 2017; 57: 423–429.

52.

Tillquist

Kutsogiannis

Wischmeyer

, et al. Bedside ultrasound is a practical and reliable measurement tool for assessing quadriceps muscle layer thickness. JPEN 2014; 38: 886–890.

53.

Durward

Amacher

Del Maestro

, et al. Cerebral and cardiovascular responses to changes in head elevation in patients with intracranial hypertension. J Neurosurg 1983; 59: 938–944.

54.

Hacker

Peters

Garkova

Ultrasound assessment of the rectus femoris cross-sectional area: Subject position implications.

West J Nurs Res 2016; 38: 1221–1230.

55.

Mayans

Cartwright

Walker

FO.

Neuromuscular ultrasonography: Quantifying muscle and nerve measurements.

Phys Med Rehabil Clin N Am 2012; 23: 133–148.

56.

Hodges

Pengel

Herbert

, et al. Measurement of muscle contraction with ultrasound imaging. Muscle Nerve 2003; 27: 682–692.

57.

Mourtzakis

Prado

Lieffers

, et al. A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care. Appl Physiol Nutr Metab 2008; 33: 997–1006.