The deep serratus anterior plane block for intensive care management of rib fractures: a cadaveric study on spread patterns

Abstract

Background

In intensive care unit management, the serratus anterior plane block is sometimes meant to be inefficient for treating rib fractures, despite its proven efficacy in thoracic surgery analgesia. This cadaveric study aimed to investigate the anatomical distribution of local anesthetic by evaluate its dissemination on fractured and non-fractured sides. The primary outcome was the distribution pattern of dye as visualised by CT imaging and confirmed by anatomical dissection.

Methods

Cadaveric study. Single-centre study conducted in the Department of Anatomy Two formalin-fixed female cadavers aged 89 and 65 years, with no prior thoracic interventions.

Bilateral deep serratus anterior plane blocks were performed under ultrasound guidance using methylene blue and contrast solution, following controlled unilateral rib fractures.

Conclusions

Both cadavers had anterolateral rib fractures. In the first cadaver, dye spread was observed toward the pleura but did not stain nerves. Dissemination was restricted due to a breast mass and thickened fascia. In the second cadaver, pleural penetration of the dye occurred on the fractured side, while the intact side showed no dye presence. Lateral cutaneous branch of the intercostal nerve, thoracodorsal nerve and long thoracic nerves staining occurred only on the non-fractured side. CT and dissection results were concordant.

Conclusions

The predictability of serratus anterior plane block efficacy may be compromised in the setting of rib fractures associated with blunt trauma due to variability in drug distribution. Given the small number of formalin-fixed cadavers used, these findings should be interpreted as preliminary anatomical observations requiring validation in fresh or Thiel-embalmed models.

Introduction

Various fascial plane blocs have entered into use for different indications in recent years [1]. One such, the serratus anterior plane block, can provide paresthesia in the ipsilateral hemothorax. The serratus anterior plane block is therefore preferred for postoperative analgesia in various operations, such as breast surgery and thoracic surgery [2]. In addition to trauma-related indications, the serratus anterior plane block has been increasingly applied in various thoracic wall procedures, including breast surgery and thoracostomy, providing effective anesthesia even in awake patients due to its reliable coverage of the anterolateral thoracic wall [3, 4]. It is also used to provide analgesia in rib fractures [5, 6]. Blunt thoracic trauma is one of the indications for intensive care admission, and rib fractures resulting from blunt thoracic trauma are associated with severe pain. Inadequate pain treatment increases pulmonary complications, and can cause patients to remain intubated for longer in intensive care, thus increasing their length of stay in intensive care, and also causes an increase in morbidity as a result [7].

Although the thoracic epidural block is regarded as the gold standard for the treatment of pain developing in association with rib fractures, due to its higher risk of complications, peripheral techniques have begun being widely used [8]. The paravertebral block or erector spinae plane block are two such examples [9]. However, the patient has to be positioned in order to perform both the epidural block and these blocks, and positioning may not be possible due to additional pathologies, such as vertebra fractures, that frequently accompany multitrauma. Unlike the erector spinae plane block or paravertebral block, which primarily target the dorsal rami or paravertebral space, the SAP block acts more superficially along the lateral thoracic wall, influencing the lateral cutaneous branches of intercostal nerves [10]. Interest has therefore grown in the serratus anterior plane block, which can be performed with the patient in the supine position, in rib fracture analgesia. Although publications have described it as efficacious, there some reported that shows poor analgesia with SAP [11]. This appears to be due to the fractures being in a posterior location, while a serratus anterior plane block targeting the lateral branches of the T2-T9 intercostal nerves may be expected to be efficacious against anterolateral rib fractures [12].

However, according to our own clinical experience, the effect of the SAP block varies in different clinical scenarios, respective of the fracture level and location. We attribute this to disturbance of the fascial planes associated with both rib fractures and blunt trauma. In the light of that hypothesis, we set out to investigate the effect of rib fractures created with blunt trauma simulation on drug spread in the deep serratus anterior plane block by means of a cadaver study.

Method

This study was designed as a single-center cadaveric anatomical investigation conducted in the Department of Anatomy, Kocaeli University Faculty of Medicine. Ethical approval for this study (Approval Number: GOKAEK-2024/18/11) was provided by the Ethics Committee of the Faculty of Medicine, Kocaeli University, İzmit, Kocaeli, Turkey (Chairperson Prof. Dr. Bülent Kara) on 21 November 2024. This report was organised in accordance with the principles of the Declaration of Helsinki and STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for descriptive and experimental research, which are based on the principles of methodological transparency [13].

Cadaver selection and preparation

Dissections were performed on two cadavers, from women aged 89 and 65. The human tissues, acquired via an institutional body donation program with agreement for exclusive use in teaching and research. The post-mortem interval before embalming was approximately 24 h. Embalming was performed using 10% formalin, and the cadavers had been stored in formalin solution for about 3 years prior to the present study. Neither cadaver had any history of surgical interventions or significant operations in the thoracic area.

Study dates

The cadaveric dissections and CT imagings were performed on 19 January 2025. Data analysis and manuscript preparation were completed between January and April 2025.

The rib fracture model: trauma simulation

In order to simulate traumatic rib fractures, controlled blunt trauma was applied to the left hemithorax of the first cadaver and to the right hemithorax of the second.

Controlled impacts were applied using a previously validated standardized method to the third to seventh ribs along the anterior axillary line by means of a special trauma device with a wooden handle and a metal head, performed by the same investigator to minimize variation [14]. The cadaver was placed on a fixed surface and prevented from moving during the impacts. A clinical blunt thoracic trauma mechanism was simulated through impacts applied at an angle of approximately 45 degrees. The fractures were confirmed by palpation and computed tomography (CT) imaging. The right hemithorax of the first cadaver and the left hemithorax of the second were left intact and employed as a control for comparative analysis.

Solution

A total of 20 mL of injectate was selected, reflecting the volume commonly used in our clinical practice and aiming to prevent excessive spread in formalin-fixed tissue, where diffusion is reduced compared with fresh cadavers. The mixture of methylene blue and iohexol enabled both visual and radiological confirmation of injectate distribution, consistent with previous cadaveric models.For the deep serratus anterior plane block, a mixture containing 1% methylene blue was obtained using 20 ml of 2% methylene blue, 17.5 ml of 0.9% saline solution, AND 2.5 ml of radiopaque dye (Iohexol, Kopaq 350 mg/1 ml, Koçsel) [15]. A final concentration of 1% methylene blue was achieved by mixing 20 ml of 2% stock solution with saline and contrast dye.

The ultrasound-guided deep serratus anterior plane block

The deep serratus anterior plane block was applied bilaterally using a high-frequency linear ultrasound (USG) probe (GE Logiq E10, 6–24 MHz, Milwaukee, WI, USA). The probe was installed on the midaxillary line in the sagittal plane, and the serratus anterior muscle and ribs beneath were visualized. The block was performed at the level of the fifth rib along the midaxillary line, where the fascial plane between the serratus anterior muscle and the underlying rib was clearly visualized. A 22 G echogenic needle (Contiplex^® A/D; B. Braun, Melsungen, Germany) was inserted using the in-plane technique, targeting the fascial plane between the serratus anterior muscle and the rib. Next, 20 mL solutions were injected into both hemithoraces to simulate the spread of local anesthetic. USG imaging was used to confirm that the injection was in the correct plane by observing hypoechogenic fascial plane expansion. All injections and dissections were performed by the same investigator to ensure consistency.

Imaging evaluation: computed tomography

Following the block application, the imaging procedures described below were applied to both cadavers in order to valuate post-injection spread:

Computed tomographic scanning

Scanning was conducted with the cadaver in the supine position using a 640-slice CT apparatus (Aquilion One, Canon). Images scanned in helical mode at 120 kV with a rotation duration of 0.5 s were examined in the soft tissue window (level: 40, width: 400) using Vital Vitrea software (Canon Group, Minnetonka, MN, USA). The radiologist assessed the images in the axial, coronal, and sagittal planes.

Anatomical dissection and stain spread analysis

Following the imaging procedures, systematic layer-by-layer cadaver dissection was carried out by a cardiovascular surgeon and anatomist experienced in thoracic wall anatomy.

• A formalin-fixed adult cadavers were used for this study.

• The cadavers were stabilized in an anatomical position to ensure optimal dissection conditions.

• The skin was incised using a scalpel, following anatomical reference points.

• Dermis and subcutaneous fat tissue were carefully separated to preserve the superficial nerve branches.

• Exposure of fascia and muscles

• Superficial fascia was removed to isolate muscle groups.

• Retractors were used to gently pull the muscle tissues aside, allowing better access to the nerves.

Isolation and anatomical examination of nerves

Thoracic lateral cutaneous branch of the intercostal nerve.

• Dissection revealed the distribution of the lateral cutaneous branches of the intercostal nerves in the thoracic wall.

• These branches were typically identified where present, originating from the intercostal nerves and supplying cutaneous sensation to the lateral thoracic area.

• Their branching points and fascial relationships were evaluated.

• Connections between these branches and the m. obliquus externus abdominis and m. serratus anterior were also examined.

Their connections to m. obliquus externus abdominis and m. serratus anterior were examined.

Nervus thoracicus longus

• Its relationship with the m. serratus anterior was thoroughly examined.

• The nerve was confirmed to provide motor innervation to the m. serratus anterior.

• Branching points and surrounding tissue interactions were identified to enhance anatomical understanding.

• Critical regions requiring surgical preservation were marked.

Nervus thoracodorsalis

• Its connection with the m. latissimus dorsi was documented in detail.

• The nerve was traced and confirming innervation to the m. latissimus dorsi.

• Once nerve examination was completed, muscle and fascial layers were repositioned to maintain anatomical integrity.

• Skin tissue was carefully preserved and closed following dissection.

• High-resolution photographs were taken throughout the dissection process to record nerve pathways.

• Findings were systematically documented in accordance with medical literature standards.

The spread was recorded in detail using high-resolution photographs during the dissection.

Data analysis

Dye spread was categorized systematically in the light of the following parameters:

1. Longitudinal and horizontal spread in the thoracic wall.

2. Penetration into the intercostal spaces and the association with the intercostal nerves.

3. The effect of the rib fractures on the dye spread was evaluated via a comparative analysis between the fractured and healthy rib sides.

4. Dissection findings were compared with the CT imaging results to determine the diagnostic value of imaging methods in evaluating injection spread.

Results

Anterolateral fractures were observed on ribs 4, 5, 6, 7, and 8 on the left side in the first cadaver. In 3D CT scan images, dye spread was found to be toward the pleura, rather than along the fascial plane (Fig. 1). Dye spread was found to be limited to the proximal parts of the intercostal nerves, but it was inconsistent between levels (Fig. 2). On the right side, a mass was observed in the breast, with adhesion in the fascial planes, and no dye spread was detected from the injection point (Figs. 1 and 3). While a more extensive methylene blue spread was expected in the right hemithorax in the first cadaver, the spread was in fact lower than that on the left. We attributed this to the mass in the right breast, which was very probably malignant and which was not encountered in the patient’s medical records. The anatomy on the right side was different to that on the left; dissection of the tissues was more difficult, the lymph nodes were clearer, and the fascia thicker. The suspected malignant mass was also confirmed by the CT images. Rib fractures evaluated as old were observed on the CT images on the same side (Fig. 4). Methylene blue spread to the pleura in the right hemithorax, and this was attributed to the mass and to the old rib fractures on that side. In addition, on the side with the breast mass, the long thoracic nerve was observed to be unstained (Fig. 5).

Fig. 1.

CT image of the first cadaver after the injection. A: Spread of the simulated side (light blue). B: Spread of the unsimulated side (dark blue)

Fig. 2.

Dye spread of the first cadaver on the left side. *Arrows show the branch of the intercostal lateral cutaneous nerves with no spread

Fig. 3.

Dye spread of the first cadaver on the right side. *Localized dye spread

Fig. 4.

CT image of the first cadaver. *Suspected malignant mass that affected the spread. White arrow shows the intrathoracic spread on the left side

Fig. 5.

The first cadaver had a breast mass and involvement of the long thoracic nerve. *Suspected malignant mass that affected the spread. White arrows show the breast mass and the long thoracic nerve

Anterolateral fractures were observed on ribs 4, 5, and 6 on the right side in the second cadaver (Fig. 6). While there was dye spread to the lateral cutaneous branch of the intercostal nerves, a small amount was observed in the pleura, although the intercostal nerves were not stained (Fig. 7). The dye spread to the long thoracic nerve (Fig. 8), and staining was also observed in the lateral cutaneous branches of the intercostal nerves arising from intercostal spaces 5–7 on the healthy left side (Fig. 9a), as well as in the thoracodorsal nerve (Fig. 9b). The dye spread to the pleura on the fractured right side in the second cadaver, but no spread was observed on the healthy side. The CT images were compatible with this.

Fig. 6.

CT image of the second cadaver after the injection. A: Blue and purple indicate the spread above the ribs. Yellow indicates the intrapleural spread.B: 3D image of the spread on the simulated side (dark blue)

Fig. 7.

Dye spread of the second cadaver on the right side. Pleural spread was observed in the pleura. However, there was no spread in the intercostal nerves

Fig. 8.

Dye spread of the second cadaver on the left side. White arrows shows the long thoracic nerve

Fig. 9.

Dye spread of the second cadaver on the left side. A: White arrows show the branch of the intercostal lateral cutaneous nerves of T5-6. B: White arrow shows the thoracodorsal nerve

However, the visibility and staining of these branches varied between specimens, consistent with known anatomical variability.

The comparative findings for both cadavers are summarized in Table 1.

Table 1.

Summary of comparative findings between fractured and non-fractured sides in both cadavers

Cadaver	Side	Fracture type	Dye spread	Pleural involvement	Nerve staining	Anatomical note
1	Left (fractured)	Ribs 4–8, anterolateral	Toward pleura, limited fascial spread	Present	None	Thickened fascia
1	Right (intact)		Minimal	Present	No staining	Breast mass restricted diffusion and old fracture
2	Right (fractured)	Ribs 4–6, anterolateral	Pleural and fascial	Present	None
2	Left (intact)		Normal fascial diffusion	None	Long thoracic, thoracodorsal stained

Pleural involvement classified as present (obvious pleural staining), mild (limited pleural contact), or none (confined to fascial plane)

Discussion

In this study, rather than the dye spreading in the fascial area, as expected, it spread directly inside the thorax in the subserratus plane block applied following rib fractures developing in association with blunt trauma. Although the drug spread to the pleura in the hemithorax in which trauma was present, variable spread was observed in the proximal parts of intercostal nerves. In addition, due to adhesions (mass) on the healthy side, the drug remained restricted to the injection point.

Previous cadaver studies have shown that in the serratus anterior plane block the drug reaches the T2-9 lateral cutaneous branch of the intercostal nerves, and the long thoracic and thoracodorsal nerves, and there is known to be no significant difference in effectiveness depending on whether the injection is performed above or beneath the serratus muscle [1]. Due to this effect mechanism it is also employed to provide analgesia in anteromedial rib fractures [16]. Indeed, there are even case reports showing it is effective in posterior fractures [17]. In their cadaver study, Johnston et al. showed that the drug can provide more effective analgesia by passing through holes in the fascia associated with the fracture to enter the intercostal nerves and can also be efficacious in posterior fractures [18]. However, since the second to sixth ribs were broken manually in that study, they were not exposed to blunt force trauma. Due to the absence of any major change in the fascial structures and muscles, the drug may also have reached the intercostal nerve while spreading as expected. In the present study, we created rib fractures by simulating blunt trauma with the application of high-energy force. We therefore think that greater deformities occurred in the fascial plane. Similarly in this study, the drug spread to pleura, inside the thorax, due to deformities in the chest wall. However, since this damage was greater than that reported by Johnston et al., it may have passed directly inside the thorax without spreading to the intercostal nerves.

This study was planned because of our experience of variable success to provide adequate analgesia with the serratus anterior plane block in patients with rib fractures associated with multitrauma in our intensive care unit, irrespective of location. Although there are similar reports regarding that the serratus anterior plane block found to be ineffective, this has been attributed to the fractures being in a posterior position [11, 12]. We think that the drug spread may also be affected by trauma-related changes in structures and emphysema. In the light of Johnston et al.’s findings, the injection point being close to the fracture line may allow the drug to spread at least to the intercostal nerve [18]. In addition, it should not be forgotten that individual factors can also affect the drug spread. Due to structural changes in the healthy side in our first cadaver, the drug remained restricted to the injection point. Such potential changes should also therefore be considered in fascial plane blocks.

It should be noted that posterior and posterolateral rib fractures, which lie closer to the paravertebral region, may produce distinct injectate spread characteristics. Although these fracture types were not simulated in the present cadaveric model, future cadaveric or in vivo studies could further clarify how fracture location influences injectate distribution and clinical analgesic efficacy. In addition, considering the need for prolonged analgesia in multiple trauma patients, it may be advisable to evaluate the efficacy of the block beforehand by means of a single-shot block application or even pre-procedure ultrasound assessment of fascial integrity to predict success before catheter insertion. Alternative techniques may be prudent considerations as a ‘Plan B.’

There are some limitations to this study. The first involves the small number of cadavers. Drug spread might usefully be investigated with more cadavers with simulated trauma. In addition, the contrast material added to visualize the spread with CT may also have affected the drug diffusion due to its higher viscosity and density compared to pure dye. Another limitation of this cadaveric study is the lack of precise information regarding the postmortem interval before embalming and the total duration of storage in formalin. An important limitation is that the volume of the dye used in the study was limited to 20 ml. While in clinic, volumes such as 30 ml and above are preferred for block application as fascial plane blocks are volume-dependent, we chose to use 20 ml because we planned this study based on our clinical observations. In our ICU, we prefer to limit the volume to 20 ml because the patients we apply this block to are multi-trauma patients and sometimes multiple blocks are needed. Therefore, we chose to use this volume for the study. However, higher volumes may yield different results. Although the impact energy was not quantitatively measured, the simulation was based on a validated protocol ensuring consistent force application. Finally, this was a cadaver study, and the tissues being non-living may also have impacted the drug spread. Although cadavers embalmed using the Thiel method are regarded as providing the best physical and functional characteristics in investigating the mechanisms of regional techniques, formalin fixation still offers valuable anatomical insight into diffusion patterns. The present research now needs to be supported by in vivo clinical studies.

Future studies including posterior fracture models, higher injectate volumes, or fresh-frozen cadavers could provide additional insight into tissue elasticity and diffusion dynamics, thereby complementing the findings of the present study.

In conclusion, it may not be possible to predict the effect of the serratus anterior plane block since it is uncertain how the drug will spread, particularly in rib fractures associated with blunt trauma

Authors’ contributions

Conceptualization: Can Aksu, Hadi Ufuk Yörükoğlu, Volkan Alparslan, Alparslan Kuş; Methodology: Volkan Alparslan, Özgür Çakır, Hadi Ufuk Yörükoğlu, Can Aksu, Abdullah Örs, Mustafa Canikoğlu; Formal analysis and investigation: Volkan Alparslan, Hadi Ufuk Yörükoğlu, Can Aksu, Özgür Çakır, Abdullah Örs, Mustafa Canikoğlu, Sevim Cesur; Writing - original draft preparation: Volkan Alparslan, Hadi Ufuk Yörükoğlu, Özgür Çakır, Mustafa Canikoğlu; Writing - review and editing: Volkan Alparslan, Hadi Ufuk Yörükoğlu, Can Aksu, Sevim Cesur, Mustafa Canikoğlu, Alparslan Kuş, Ercüment Çiftçi, Tuncay Çolak; Supervision: Alparslan Kuş, Ercüment Çiftçi, Tuncay Çolak.

Funding

This research received no external funding.

Data availability

The dataset, including CT images and dissection figures, will be made available upon reasonable request via e-mail.

Declarations

Ethics approval and consent to participate

Ethical approval for this study (Approval Number: GOKAEK-2024/18/11) was provided by the Ethics Committee of the Faculty of Medicine, Kocaeli University, İzmit, Kocaeli, Turkey (Chairperson Prof. Dr. Bülent Kara) on 21 November 2024. As this study did not involve patients, patient data, or any clinical interventions the need for written informed consent was waived by the ethics committee. The study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

Hadi Ufuk Yörükoğlu and Can Aksu are members of the Editorial Board of BMC Anesthesiology. All other authors declare that they have no competing interests.