The decision to ventilate and eventually to extubate is an everyday practice in intensive care units. Despite its usefulness, prolonged use of mechanical ventilation (MV) results in several hazards. Conversely, premature separation of MV leads to reintubations with serious cardiorespiratory consequences and increased mortality.1 Failed weaning is a complicated, multifaceted event attributed to either direct respiratory or extraneous causes.2 Consequently, the timing for separation from mechanical ventilation must be thoroughly assessed; although various screening techniques have been proposed to predict weaning outcomes, including rapid shallow breathing index (RSBI), perfusion index, and dyspnea intensity.3–5 None has proven to be optimal thus far. Notably, ultrasound is a safe, noninvasive, and readily available bedside tool that provides immediate and dynamic information, making it particularly valuable in the critical care setting for assessing readiness for extubation. In this context, respiratory ultrasound evaluation emerged as an accurate, bedside, non-invasive, and easily repeatable tool, including assessment of lung parenchyma, diaphragm, and extra-diaphragmatic accessory muscles. Moreover, respiratory ultrasound guides the diagnosis and management of acute respiratory failure, improving clinical decision-making and patient outcomes.6 Nonetheless, there are some challenges, including inappropriateness for patients with elevated abdominal pressure, difficulty conducting left-side examinations, and a lack of predictive capability in specific populations.7–12 Furthermore, diaphragmatic assessments may be influenced by the administration of sedative medications and the existence of pain.13–15 There is a well-known inverse relationship between the diaphragm and accessory respiratory muscles; hence, in the case of diaphragmatic dysfunction, accessory respiratory muscles increase their activity to keep adequate tidal volume.7,8,16–18 The diagnostic accuracy of accessory respiratory muscles, such as the parasternal intercostal and abdominal muscles, in predicting weaning outcomes is becoming increasingly evident.16,19,20 The anterior scalene muscle is a superficial accessory inspiratory muscle that originates from the transverse processes of cervical vertebrae C3 to C6 and is inserted into the first rib.21 The scalene muscle electromyography, as suggested by recent evidence, has a positive correlation between electrical activity and dyspnea intensity.22 We hypothesized that the activity of the scalenus anterior muscle increases in patients with increased workload or diaphragmatic weakness; therefore, it can be used as a marker for weaning outcomes. To the best of our knowledge, no previous studies have investigated the role of scalenus anterior thickening as a predictor of weaning outcomes. Therefore, we aimed to evaluate the feasibility and reproducibility of scalenus anterior thickening as well as its ability to predict weaning outcomes.

This prospective observational study was carried out at a university hospital after approval from the institutional research ethics committee (MD-155-2024) and concordance with Helsinki's declaration. This study was conducted over six months from July 2024 to December 2024 and reported by the Strengthening Reporting of Observational Studies in Epidemiology (STROBE) guidelines. The patient or their next of kin signed an informed consent form before the commencement of the study.

This study was divided into two parts: sub-study A, which evaluated the feasibility and reproducibility of the right scalenus anterior thickening in healthy subjects, and sub-study B, which included sixty-six critically ill patients undergoing spontaneous breathing trials (SBT).

Sub-study A: Two experts in respiratory muscle ultrasonography conducted an exploratory study that included 21 healthy volunteers to determine the feasibility and reproducibility of scalene muscle evaluation. Each examined the right scalenus anterior thickness during inspiration and expiration, and the scalene thickening fraction was determined. All patients were given a thirty-minute rest before evaluation. Observer 1(MH) conducted two sets of examinations separated by a 10-minute delay; after 20 min, observer 2 (LM) performed comparable examinations. Both observers were blind to the measurements of the other. Intra and interobserver reliability were checked for right-side examinations.

Sub-study B: Included sixty-six adult critically ill patients who were mechanically ventilated for 24 h or more and were eligible for SBT. Patients fulfilling readiness for weaning criteria (Supplementary file) were subjected to SBT for 30 min using zero-pressure support and zero positive end-expiratory pressure, since our main objective was to evaluate the patient's intrinsic respiratory effort without the influence of ventilatory assistance as a confounding factor. Patients with neuromuscular disorders, tracheostomy, and limited ultrasound views (poor acoustic window, subcutaneous emphysema, surgical dressing, or central venous catheter at the site of examination) were not eligible. Failed weaning was defined as failed SBT or reintubation within 48 h. Criteria for SBT failure included increased heart rate > 140 beats/minute or respiratory rate > 35 breaths/min, significant blood pressure fluctuations, specifically systolic blood pressure (SBP) > 180 mmHg, SBP < 90 mmHg, and /or a deviation exceeding 20% from baseline values and signs of increased accessory muscle activity and sweating. As per our institutional post-extubation protocol, prophylactic application of non-invasive ventilation (NIV) was indicated for patients identified as high-risk, specifically those with chronic obstructive pulmonary disease, an APACHE II score greater than 12, a duration of mechanical ventilation exceeding seven days, and age over 65 years. Importantly, NIV was employed exclusively as a prophylactic measure and not as a rescue intervention in cases of post-extubation respiratory failure. This approach was intended to prevent confounding extubation failure rates.

An experienced operator (AE), who completed more than 50 similar examinations before the commencement of the study, performed the diaphragmatic and scalene muscle ultrasound. The operator was blinded to the clinical data and not involved in the weaning process. Examinations were conducted using the Versana Essential device (GE Medical Systems Co., Ltd., China) at mid-trial (15 min after initiation of SBT) while the patient was semi-sitting. For scalene muscle evaluation, a high-frequency linear transducer (L6-12-RS, 4–16 MHz) was placed horizontally at the level of the cricoid cartilage with the neck rotated to the opposite side; at this point scalenus anterior muscle can be visualized clearly below the lateral border of sternocleidomastoid muscle, brachial plexus between the anterior and middle scalene muscles, and finally internal jugular vein below scalenus anterior muscle. The M mode was then switched on in the middle of the muscle with the sweep speed adjusted to a minimum to allow multiple breaths to be obtained on the same screen (Figs. 1, 2). Inspiratory and expiratory thickness was measured as an average of 3 breaths; subsequently, the thickening fraction was calculated as (inspiratory thickness – expiratory thickness)/expiratory thickness and expressed as a percentage. The diaphragmatic examination was in concordance with a recent consensus statement23 and current literature9,10,17 using a curved probe (4C-RS, 2–5 MHz) placed at the 10th or 11th intercostal at the right anterior axillary line and the left midaxillary line for each hemidiaphragm assessment, patient was then asked to take maximum breath, subsequently M mode was switched on to measure diaphragmatic excursion as an average of 3 successive breaths. The main outcome variable in this study was the ability of the scalenus anterior thickening fraction (SA-TF) to predict failed weaning (failed SBT and reintubation); other outcomes were the ability of RSBI to predict weaning outcomes, and finally, the correlation between diaphragmatic excursion and scalene thickening fraction.

Figure 1.

Scalenus anterior thickening.

A-probe position for right side examination, B-probe position for left side examination.

C-ultrasound view of right scalenus anterior, D-ultrasound view of left scalenus anterior.

1-sternomastoid muscle, 2-anterior scalene muscle, 3-middle scalene muscle.

4-brachial plexus, 5-inspiratory thickness, 6-expiratory thickness.

Figure 2.

Illustrative diagram for the scalene muscle ultrasound.

CA: carotid artery, Exp: expiration, IJV: internal jugular vein, Ins: inspiration.

Sub-study A: An Exploratory study included 21 healthy subjects, with 12 being males. The mean age for included subjects was 34.1 (9.7) years, and BMI was 25.1 (2.1). The inspiratory thickness of the right anterior scalene muscle was significantly greater in males (14.2 ± 1.7 mm) compared to females (11.6 ± 1.3 mm; P < .05). However, the absolute difference was relatively small, and given that our primary focus was on the thickening fraction rather than absolute thickness, this sex-based variation was not considered central to outcome prediction (Table 1). Intra-class correlation coefficient 0.94, 0.95 for observers 1 and 2, respectively. With interclass correlation coefficients of 0.79 and 0.8 for inspiratory and expiratory muscles, respectively (Table 2).

Sub-study B: Seventy-five critically ill patients were screened for eligibility; 7 patients were excluded, and 66 patients were available for final analysis; 41/66 (43%) showed successful weaning, 18/66 failed SBT, and 7/66 were reintubated (Fig. 3). Mean age and days of mechanical ventilation were higher in patients with reintubation. (Table 3) Patients who successfully weaned showed higher diaphragmatic excursion and lower SA-TF (Table 4). Patients’ comorbidities were summarized in Table 4. It is noteworthy that reintubated patients showed a significantly lower expiratory muscle thickness (Supplementary Table). Receiver operating characteristics (ROC) curve analysis for SA-TF revealed good accuracy for predicting SBT failure with an AUC (95% CI) of 0.92 (0.82−1) and 0.94 (0.84−0.98) for the right and left sides, respectively. Added to that, SA-TF could predict reintubation with an AUC of 1.00 and 0.99 for the right and left sides, respectively (Fig. 4) (Table 5). SA-TF < 10% can rule out weaning failure with 100% negative predictive value. Rt SA TF's reported cut-off values for predicting failed SBT and reintubations were 21% and 12%, respectively. While that of Lt SA TF was 10% (Table 5). Interestingly, there is a strong negative relationship between mean diaphragmatic excursion and mean scalene thickening (Fig. 5).

Figure 3.

Patient enrollment.

ICU: intensive care unit, SBT: spontaneous breathing trial.

Figure 4.

Receiver operating characteristics curve.

A-for the ability to predict failed SBT, B-for the ability to predict reintubation.

DE: diaphragmatic excursion, RSBI: rapid shallow breathing index, SBT: spontaneous breathing trial, TF: thickening fraction.

Figure 5.

Correlation between diaphragmatic excursion and scalenus anterior thickening fraction.

Notably, univariate analysis showed significant associations with several variables, including age (OR 1.05, 95% CI 1.01–1.09, P = .016), days of mechanical ventilation (OR 1.30, 95% CI 1.05–1.61, P = .015), diabetes mellitus (OR 5.51, 95% CI 1.86–16.29, P = .002), ischemic heart disease (OR 7.58, 95% CI 1.43–40.19, P = .017), and PaCO₂ (OR 0.90, 95% CI 0.83–0.99, P = .029). Increased heart rate (OR 1.07, 95% CI 1.02–1.11, P = .006) and elevated RSBI (OR 1.10, 95% CI 1.05–1.16, P < .001) were also significant. Importantly, a lower M-DE (OR 0.51, 95% CI 0.36–0.73, P < .001) and higher M-SA TF (OR 1.37, 95% CI 1.18–1.59, P < .001) were strongly associated with weaning failure (Table 6).

The current study demonstrated that the scalenus anterior thickening is a feasible and reproducible tool with good intra- and inter-observer reliability. Additionally, the scalene thickening fraction can accurately predict weaning outcomes. Notably, sex-based differences in scalene muscle thickness were observed; however, these are physiologically expected and unlikely to influence outcomes, as the thickening fraction, rather than absolute values, was used in our analysis.

This study is the first to evaluate anterior scalene muscle thickening as a predictor of weaning outcomes, offering novel insights into the potential role of superficial neck muscles in respiratory assessment. Given their accessibility and ease of ultrasound evaluation, this approach may be particularly advantageous in patients with abdominal or thoracic surgical interventions, or in those with increased abdominal distension, where traditional diaphragmatic imaging may be limited. Despite being a widely used tool to predict weaning outcomes, diaphragmatic ultrasound showed some challenges, including limited ultrasound views, not ideal for patients with abdominal surgery at or close to the site of assessment, difficulty in evaluating the left side, and being highly dependent on the level of pressure support.9,10,17,23,25,26 Therefore, accessory inspiratory and expiratory muscle evaluation has emerged as an alternative tool with comparable diagnostic accuracy.7,16,19 We report a strong negative relationship between diaphragmatic excursion and scalenus anterior thickening fraction; previous literature reported a similar negative relationship between diaphragmatic function and accessory respiratory muscles.16,19,20 This occurs because, in the case of diaphragmatic dysfunction, the accessory respiratory muscles increase their activity to compensate for both the increased workload and the diaphragmatic dysfunction, thereby maintaining adequate tidal volume. In line with our results, a recent physiologic study reported increased scalene electric activity with increased respiratory workload.22 The performance of abdominal expiratory muscles may be restricted due to recent abdominal surgery or elevated intra-abdominal pressure; also, the assessment of parasternal intercostal muscles is sensitive to the level of pressure support.27

This study was conducted at a single center with a relatively small sample size. Second, we did not examine the lung parenchyma; therefore, future studies are required to evaluate the combined evaluation of scalenus muscle thickening and lung ultrasound scores. Third, anatomical factors, including central lines, neck edema, and excessive soft tissue, may limit visualization of the scalene muscle. Finally, although univariate analysis identified key predictors of weaning failure, multivariable regression was not feasible due to quasi-separation arising from a small sample size and low event rate. This statistical constraint limits the ability to adjust for confounders and accurately assess independent effects. Larger future studies are needed to support comprehensive modeling and validate these associations.

The anterior scalene muscle thickening fraction (SA-TF) appears to be a feasible and reproducible ultrasound parameter, with good intra- and inter-observer reliability. In critically ill patients ventilated for ≥24 h, SA-TF showed promising predictive value for weaning outcomes. Although an SA-TF < 10% was associated with a high negative predictive value for weaning failure, these findings should be interpreted cautiously pending external validation and adjustment for confounding variables in future studies.

M.H.: This author helped in study design, data acquisition, and manuscript review. L.M.: This author helped in study design, data acquisition, and manuscript review. A.S.: This author helped in the conception of the idea, study design, analysis of the data, and drafting the manuscript. A.E., W.H.: These authors helped in study design, data acquisition, and manuscript review. All authors approved the manuscript and agreed to be accountable for all aspects of the work.

Ethical approval for this study was provided by the institutional research board of Cairo University Hospitals (MD-155-2024).

Informed consent was obtained from each patient or their next of kin before participation.

Not applicable.

Grammarly and Quillbot software were used for English editing.

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data regarding this study is available from the corresponding author upon reasonable request.