Ultrasound assessment of rectus femoris and anterior tibialis muscles in young trauma patients

We enrolled exclusively young trauma patients with an injury severity score(ISS) exceeding 25, who were admitted to our ICU within few hours after the injury. We recruited only well-nourished, previously healthy subjects, aged 18–59 y.o, with no past history of nutritional problems, chronic use of drugs nor orthopedic issues (such as skeletal fractures or immobilization) in the previous 2 years. Trauma patients not fulfilling this criteria were not considered for the enrollment. Exclusion criteria were: relevant comorbidities (renal, liver or heart disease or COPD), previous immune abnormalities (including treatment with corticosteroids), neuromuscular disease, past or recent history of cancer.

For each patient, demographics and clinical data were recorded: age, height, weight, body mass index (BMI), ISS, APACHE II score, Sequential Organ Failure Assessment score (SOFA), Glasgow Coma Scale (GCS)—both total and motor (GCS-M)—Glasgow Outcome Scale (GOS), number of days on mechanical ventilation, ICU length of stay, incidence of secondary infections, daily provision of calories and protein, blood levels of albumin, total protein and creatinine, blood urea nitrogen (BUN) and other laboratory data. Among mechanically ventilated patients, weaning was classified as simple, difficult or prolonged, as previously described [14]. Infections were classified according to definitions by the Center for Disease Control and Prevention [15].

Enteral feeding was started as soon as the patient was hemodynamically stable and fully resuscitated, usually within 24 h. Our nutritional target was to achieve a minimal protein intake of 0.8 g/kg/day within day 5 after admission. Patients who did not reach this target for any reason (gastrointestinal intolerance or contraindication to enteral feeding or repeated forced suspensions of enteral feeding because of multiple surgical procedures) were excluded from the final analysis. We used either a standard feeding formula (1.5 total kcal/ml, protein 60 g/L) or a high-protein formula (1.35 total kcal/ml, protein 75 g/L): These different regimens were not assigned by randomization, but they were the result of a change of nutritional policies in our ICU during the period of the study; though, the two groups of patients—nourished by a standard formula (SF) or nourished by a high-protein formula (HPF)—were similar in terms of age, Injury Severity Score (ISS), Acute Physiology and Chronic Health Evaluation (APACHE II score), height and weight. Tube feeding was started at a very low rate (10–20 ml/h) and increased at each 24-h interval as tolerated, in order to meet a minimal protein intake of 0.8 g/kg/day. The patient’s intolerance to feeding was defined on the basis of clinical signs (abdominal distention, vomiting, increase in serum lactate, high gastric residual volume).

An ultrasound (US) evaluation of the rectus femoris (RF) and anterior tibialis (AT) muscles was performed in all patients at day 0 (within 24 h from trauma), 5, 10, 15 and 20. We used an US device with a 5- to 7.5-MHz linear probe (Esaote MyLab). According to a previously described methodology [13, 16], skeletal muscles were evaluated by US scan, collecting both quantitative and qualitative data. The transducer was placed perpendicular to the long axis of the muscle (i.e., perpendicular to the major axis of the limb), at 3/5 of the distance between the anterior superior iliac spine and the superior border of patella (i.e., about 15 cm from the patella) for the RF and 5 cm below the peroneal head for the AT muscle. The measurement points were marked with indelible ink to ensure day-to-day consistency and facilitate subsequent measurements. Visualization was consistently obtained, while the patients were supine with both legs in passive extension. Two measurements were taken for each muscle in each leg. In the presence of lower limb fractures, measurements were taken on the contralateral leg only. Excess contact gel was applied so to minimize underlying soft tissue distortion. One operator only (specifically trained in muscle ultrasonography) performed all measurements. US settings (depth, gain and focus) were standardized for both RF and AT examinations. After freezing the US image, quantitative parameters were recorded for both muscles: anterior–posterior diameter (AP diam); lateral–lateral diameter (LL diam); and cross-sectional area (CSA) (computed from the perimetral contour of the muscle section). The value of CSA is considered to be proportional to the total mass of the skeletal muscle [13, 16]. We also recorded one qualitative parameter—echogenicity—that was expressed according to the Heckmatt Scale [17]. Echogenicity of normal muscle is expected to be 1 on this scale. Increased echogenicity is usually regarded as an index of myofibers depletion [18].

Qualitative data are expressed as number of events (%) and continuous data as median [interquartile range]. Difference in the distribution of qualitative variables in SF versus HPF group was investigated with the Chi-square test or Fisher’s exact test, as appropriate. Difference in the distribution of quantitative and ordinal variables in SF versus HPF group was assessed with the Mann–Whitney test. In the overall population, the significance of changes in the quantitative variables over time was determined with the one-way ANOVA test for repeated measures; paired comparisons between 2 consecutive timepoints were then analyzed with the Wilcoxon sum of ranks test. The effect of SF versus HPF on the change in the quantitative variables over time was determined with the two-way ANOVA. Paired comparisons between the distribution of quantitative variable in SF versus HPF group at each time point were also assessed with the Mann–Whitney test.

Statistical analysis was conducted with SPSS 20.0.

In ICU patients, the daily amount of muscle loss—as estimated by US—is reported to range between 6 [20] and 12.5% between day 1 and 7 [4]. Muscle wasting correlates with the ICU length of stay [16] and can be predictive of long-term functional disability [21]. Several factors contribute to muscle wasting over the course of the critical illness, both in the acute and chronic phase: inflammation, neuroendocrine stress response, immobilization, impaired microcirculation and denervation (in the acute phase); infections, nutritional deficiency, hyperglycemia, drugs [22] (in the late phase). Other predisposing factors are: age, baseline muscle function, nutritional status, comorbidities (COPD, renal and heart disease, cancer) [22, 23]. Finally, some data suggest that also parenteral nutrition may worsen muscle function [24].

A retrospective study demonstrated a correlation between hospital mortality and skeletal muscle mass, as estimated by abdominal CT scan [2]: this is particularly evident in trauma patients [25], but apparently not in ICU patients with acute lung injury [26].

Ultrasound has been used to rate the loss of skeletal muscles in patients with orthopedic trauma, COPD, cancer and neuromuscular disorders [13, 27, 28]; indeed, it appears as an emerging field of interest in ICU. Ultrasonography is more accurate than anthropometric measurements and has been shown to closely correlate with the data obtained by MRI and CT scan [19, 29], with the advantage of being less expensive, less time-consuming and safer, since it does not imply radiation exposure. Though some studies have shown a good intra-rater and inter-rater reliability for US measurement of muscle CSA or thickness in adult critically ill patients [4, 30], the matter is still somehow controversial [31, 32].

All of our trauma patients (100%) experienced severe muscle mass loss, as estimated by CSA. Almost half (45%) of RF muscle mass was lost by day 20 with the greatest reduction (21%) occurring after day 15. In a previous work [4], a 17.7% reduction in RF cross-sectional area was shown in a group of mixed ICU patients from day 1 to day 10, with the major loss occurring during the first 7 days. We found a less important reduction in AT cross-sectional area (22%) by day 20.

The exact underlying mechanisms of dissimilar magnitudes of losses in different muscle groups are still unknown. In both rodent and human models, the rate and magnitude of muscle loss seem to depend on both muscle type and degree of inactivity [33, 34]. In experimental and clinical models of lower limb immobilization, muscle loss is greater in the extensor muscles (soleus and gastrocnemius). This is consistent with the greater muscle loss we report in RF (extensor muscle) as compared to AT (flexor muscle). Also, RF is a power muscle made up predominantly of type II fast-twitch fibers, while AT muscle composition is mainly made of type I slow-twitch fibers [35]. The preferential loss of a certain kind of muscle fibers might be a crucial determinant of long-term outcome, especially in the development of ICU-acquired weakness and in success of physical rehabilitation. Muscle weakness is usually symmetric and predominates in the proximal part of the limbs (shoulders and ankles) [36]. Laboratory model of ischemic injury has shown that muscle with predominance of fast-twitch fibers had significantly greater necrosis than those richer in slow-twitch fibers [37]. Immobilization studies have shown a preferential loss of type II fibers and conversion of fiber typing from type I to type II in postural muscles [38].

Several studies have demonstrated that pathological muscle changes (such as fatty infiltration, atrophy and intramuscular fibrosis) can be detected by ultrasound. Alteration of muscle echogenicity may be ascribed to muscle edema (in the early phase) but also to fibrosis and fatty degeneration (in the late phase). These latter findings may be an indicator of quantitative loss of muscular myofibers and disruption of muscle architecture and may correlate with impaired muscle function [18]. Since edema cannot alter the bone signal in contrast to fibrous tissue, changes in muscle echogenicity are related to the fibrous tissue content and with specific structural damage in muscle architecture as seen with muscle biopsies [4] or with muscle magnetic resonance imaging [18]. Structural muscle changes detected by the increased echogenicity have been correlated with measures of muscle strength and function [39].

In our study, echogenicity was quantified by the Heckmatt Scale (Table 5), previously used in the critically ill setting by Grimm [18]. A higher grade of echogenicity with reduced bone signal correlates with the severity of myopathy [17]. Of course, the use of a semiquantitative method such as the Heckmatt Scale may be biased by observer dependency and technical misinterpretation in contrast to objective, user-independent algorithms for image analysis such as computer-assisted quantitative grayscale analysis [40]. Though, the Heckmatt Scale has the relevant advantage of being a rapid and inexpensive bedside technique that can be easily used in the intensive care setting.

Changes in muscle architecture have been documented in previous studies on ICU patients and are associated with increased length of stay in ICU [4, 16, 18, 27, 29]. Changes in muscle echogenicity [4, 18] suggest an alteration of myofibers content, secondary to edema from capillary leak or inflammation. These data are confirmed by muscle biopsies on day 1 and 7 after ICU admission, showing muscle necrosis and macrophage infiltrate [4].

In our study, we found a progressive increase in both RF and AT echogenicity from day 5 on. An alteration of echogenicity was already evident in AT muscle soon after admission and tended to increase in the following weeks. The early alteration of echogenicity in AT but not in RF may have many explanations; probably, post-traumatic edema was more pronounced in the muscles of the lateral/posterior part of the limb (AT) than in the anterior area (RF).

An optimal provision of energy and protein has been regarded as an important factor improving the patient’s chance of survival and satisfactory clinical outcome [41]. Provision of an optimal amount of protein has been shown to improve the rate of protein synthesis in tissues with rapid turnover, though it did not reduce the catabolic response to injury [42, 43]. In one recent multicenter study [44], provision of at least 80% of the prescribed protein (i.e., 1 g/kg/day) reduced mortality in a ICU population. In patients on parenteral nutrition, protein delivery may be more important than caloric support in terms of short-term outcome [11].

In our study, we found no difference in muscle mass loss or in muscle echogenicity between patients fed with standard (SF) versus high-protein formulas (HPF): though, no conclusion can be drawn in this regard, since the study was not designed or powered to verify such hypothesis. Nonetheless, this finding may be consistent with previous studies showing that depletion of lean body mass, particularly skeletal muscle, is not influenced by nutritional support [45‐47] and with studies showing an inverse correlation between the amount of protein delivery and the cross-sectional area of RF [4].

Immobilization and inflammation—rather than inadequate nutritional support—might be major determinants of loss of muscle. Inactivity is a potent stimulus to muscle protein breakdown and activation of the ubiquitin–proteasome pathway of proteolysis [48]. Immobility of limbs is quite common in ICU patients and is related to bed rest and sedation. Acute and chronic activation of inflammatory pathway is another potent stimulus for proteolysis [49]. Though, in our study we did not measure any index of inflammatory activity and we could not verify such contention.