Article Text

Download PDFPDF

Transversus thoracis muscle plane block in cardiac surgery: a pilot feasibility study
  1. Satoru Fujii1,
  2. Matthew Roche1,
  3. Philip M Jones1,2,
  4. Deepti Vissa1,
  5. Daniel Bainbridge1 and
  6. Jian Ray Zhou1
  1. 1 Department of Anesthesia and Perioperative Medicine, London Health Sciences Centre, Western University, London, Ontario, Canada
  2. 2 Department of Epidemiology and Biostatistics, Western University, London, Ontario, Canada
  1. Correspondence to Dr Satoru Fujii, Department of Anesthesia and Perioperative Medicine, London Health Sciences Centre, Western University, London, ON N6A 3K7, Canada; satoru.fujii{at}


Introduction Cardiac surgery patients often experience significant pain after median sternotomy. The transversus thoracis muscle plane (TTP) block is a newly developed, single-shot nerve block technique that provides analgesia for the anterior chest wall. In this double-blind pilot study, we assessed the feasibility of performing this novel block as an analgesic adjunct.

Methods All patients aged 18–90 undergoing elective cardiac surgery were randomized to the block or standard care control group on admission to the intensive care unit after surgery. Under ultrasound guidance, patients in the block group received the TTP block with 20 mL of either 0.3% or 0.5% ropivacaine bilaterally, based on weight. The control group did not receive any injections. All blocks were performed by a single anesthesiologist, and data collection was performed by blinded assessors. The primary feasibility outcomes were rate of recruitment, adherence, and adverse events. The rate of recruitment was defined as the ratio of patients giving informed consent to the number of eligible patients who were approached to participate. Secondary outcomes included 12-hour and 24-hour Numeric Rating Scale (NRS) pain scores, 24-hour hydromorphone and acetaminophen requirements, time to extubation, time to first opioid administration, and patient satisfaction (on a yes/no questionnaire) at 24 hours.

Results Twenty patients were approached for this study and 19 were enrolled. Eight patients received the intended intervention in each group. The recruitment rate was 95% of all approached eligible patients, and the adherence rate to treatment group was 94%. There were no block-related adverse events. The mean (SD) NRS pain scores at rest were 3.3 (3.2) in the block group vs 5.6 (3.2) in the control group at 12 hours. At 24 hours, the pain scores were 4.1 (3.9) vs 4.1 (3.3) in the block and control group, respectively. The mean (SD) 24-hour hydromorphone administration was 1.9 (1.1) mg in the block group vs 1.8 (0.9) mg in the control group.

Discussion The TTP block is a novel pain management strategy poststernotomy. The results reveal a high patient recruitment, adherence, and satisfaction rate, and provide some preliminary data supporting safety.

Trial registration number NCT03128346.

  • Transversus thoracis muscle plane block
  • TTP
  • pain management
  • cardiac surgery
  • regional anesthesia
  • sternotomy

Statistics from


Cardiac surgery patients can experience significant pain after a median sternotomy, and standard treatment with oral and intravenous analgesics is not always adequate.1–3 A prospective study of 705 cardiac surgery patients reported pain scores ranging between 5.3 and 6.5 out of 10 with deep breathing and coughing on postoperative day 1.3 Inadequate pain relief after surgery can lead to sustained postoperative pain; based on previous studies, chronic chest pain poststernotomy is a serious problem for 17%–88% of patients.4–6 According to one study,1 the mid-sternum is the source of maximal pain in most patients postoperatively. The sites of vein harvest and chest tube insertion also contribute to postoperative pain but to a lesser degree. Although some attempts to introduce neuraxial and paravertebral block techniques have been made in the past, many factors in the cardiac surgery patients have prevented regional techniques from being widely used, including hemodynamic instability and postoperative coagulopathy and anticoagulation.

The transversus thoracis muscle plane (TTP) block is a newly developed regional block technique that provides analgesia for the anterior chest wall, covering the intercostal nerves from T2 to T6.7–10 Ueshima and Kitamura8 first described the clinical use of the TTP block in 2015 for an 86-year-old patient with severe heart dysfunction undergoing segmental breast resection. The combination of the TTP and Pectoralis (Pecs) block enabled successful awake breast resections. Subsequent case reports and a randomized controlled trial have demonstrated the superior analgesic effects of the TTP and Pecs block for mastectomy over the Pecs block alone.9–11 Recently, a case series of two patients showed bilateral TTP blocks could provide effective analgesia for patients undergoing median sternotomy.12 Due to the potential for better pain control poststernotomy, we conducted a double-blind pilot study to assess the feasibility of performing this novel TTP block as an analgesic adjunct after cardiac surgery. This pilot study was conducted in accordance with the definitions and protocol described in a published tutorial on pilot studies.13 The purpose of this study is to assess the feasibility of the TTP block in postcardiac surgery patients.


Patients and design

This study was a single-center, prospective, randomized, double-blind, controlled, feasibility study.

All patients aged 18–90 undergoing elective cardiac surgery requiring median sternotomy were screened on the day of surgery for inclusion in the study. Patients undergoing saphenous vein and radial artery harvests were included in this study despite the possible confounding effects on postoperative pain. As this feasibility study will serve as a reference for future clinical trials, it was necessary to include the above-mentioned patient population.

The exclusion criteria were patient refusal, lack of informed consent, emergency or non-median sternotomy surgery, patients with a diagnosis of cognitive impairment, known local anesthetic (LA) allergy, pre-existing major organ dysfunction including hepatic or renal failure, and left ventricular ejection fraction <30%, significant psychiatric illnesses including schizophrenia, bipolar disorder, uncontrolled anxiety, or depression, chronic opioid use of greater than 15 mg oral morphine equivalents daily, peripheral neuropathy, pregnancy, coexisting hematologic disorders, and a language or reading barrier.

In this feasibility study, due to a lack of previous studies, a convenience sample size of 20 was used. In order to facilitate consistency in the recruitment process, a recruitment rate of one patient per week was targeted.

Surgery, anesthesia, and analgesia

In the operating room, anesthesia was induced with propofol 0.5–1 mg/kg, midazolam 0.05–0.1 mg/kg, and fentanyl 2–5 µg/kg. Rocuronium 0.6–1.0 mg/kg was administered to facilitate orotracheal intubation with a cuffed tracheal tube. Anesthesia was maintained with isoflurane 0.5%–1.0% in air and oxygen. Patients and the intraoperative team were unaware of the study group allocation since the randomization was performed postoperatively. No changes to intraoperative care were mandated due to this study. The administration of intraoperative opioids was left to the discretion of the attending anesthesiologists. Postoperative analgesic medication administration was directed by local intensive care unit (ICU) standard practice. Our study protocol recommended opioids be considered when patients indicated pain scores on the Numeric Rating Scale (NRS) >4.

Randomization and blinding

Within the first 2 hours of admission to the cardiac surgery ICU, with patients still intubated, randomization was performed at the bedside to either TTP block or standard care group using the Research Electronic Data Capture (REDCap) software (V.8.1.2; Vanderbilt University Medical Center, Tennessee, USA). After confirming that the patient was stable from a hemodynamic and coagulation standpoint, the investigator administered bilateral TTP blocks under dynamic ultrasound guidance for patients in the block group. The control group received an ultrasound assessment of the TTP in the anterior chest wall but did not receive any injections. Outcome assessors were blinded to group allocation by ensuring all ICU bay curtains were drawn during the procedure, and surgical bandages were applied to the site of injection in all patients.


All blocks were performed using real-time ultrasonography (US) with a high-frequency linear transducer (SonoSite Edge II, Fujifilm SonoSite, Washington, USA). Patients were positioned supine with chest exposed and monitored by beat-to-beat arterial tracing and standard American Society of Anesthesiologists (ASA)monitors. No intravenous medications were required for block performance as the patients were sedated for mechanical ventilation postsurgery.

After determining the anterior T4–T5 interspace using US (figure 1), the US probe was placed in the longitudinal plane 1 cm lateral to the sternal border (figure 1).14 A parasternal sagittal view of the internal intercostal muscle and the transversus thoracis muscle between the fourth and fifth rib was visualized above the pleura (figure 1). A 22-gage, 80 mm Pajunk needle (Pajunk Medical Produkte, Germany) was inserted inplane to the transducer with the tip of the needle located in the TTP between the internal intercostal and transversus thoracis muscles (figure 1). After excluding intravascular and intrapleural placement, LA was administered in 5 mL aliquots with intermittent aspiration (figure 1).

Figure 1

(A) The injection site (*) is at the intersection of the T4/T5 rib space and 1 cm lateral to the sternal border, just adjacent to the sternal dressing. (B) The ultrasound transducer is oriented longitudinally along the sternal border, along the dotted line. (C) The needle is inserted inplane to the transducer. (D) The TTP can be seen between the fourth and fifth rib shadows, between the internal intercostal muscle (IIM) and the transversus thoracis muscle (TTM) which lies above the pleura. (E) The needle tip can be visualized in the TTP. (F) Local anesthetic (LA) can be visualized in the TTM during and post block. PM, pectoralis major; TTP, transversus thoracis muscle plane.

Figure 2

Flow diagram of patients enrolled in the study. EF, ejection fraction.

Patients in the block group received bilateral TTP blocks with 20 mL of 0.3% (patients weighing <75 kg) or 0.5% ropivacaine (patients weighing ≥75 kg), based on a weight cut-off of 75 kg. Patients in the control group had ultrasound scanning of the chest and surgical bandages placed at the site of ‘injection’ bilaterally, but did not receive a block. After block administration, the patients were monitored for LA toxicity, hemodynamic instability, and allergic or unexpected adverse reactions for 20 min. All blocks were performed by a single anesthesiologist (SF). The time required to perform the blocks bilaterally ranged from 10 to 15 min.

Outcomes and assessments

The primary feasibility outcomes of this pilot study were the rate of recruitment, rate of adherence, and rate of adverse events up to 48 hours postsurgery. The rate of recruitment was defined as the ratio of patients giving informed consent to the number of eligible patients who were approached to participate. The rate of adherence was defined as the percentage of patients who receive the randomized intervention. Possible complications include pneumothorax, LA toxicity, allergic reaction, and pain, hematoma, and infection at the block site. The presence of pneumothorax was ruled out by a postoperative chest X-ray read by an independent radiologist.

Secondary outcomes included 12-hour and 24-hour NRS score (0=no pain, 10=worst pain imaginable), 24-hour hydromorphone and acetaminophen administration, time to extubation, time to first opioid dose in the ICU, non-steroidal anti-inflammatory drug and antiemetic use, and patient satisfaction regarding postoperative pain management (yes/no) at 24 hours. Time 0 for data collection commenced at the time of randomization.

Data collection was performed by the primary ICU nurse or a research assistant, who were both blinded to patient allocation. De-identified data were stored in a secure REDCap database hosted at Lawson Health Research Institute.

Statistical analysis

Descriptive statistics were calculated and summary statistics for the outcomes studied were compiled. Since this was a pilot study, no inferential statistics were performed.


From May to October 2017, 22 patients were screened for this study. Two patients were excluded based on exclusion criteria, and one patient declined participation. Of the 20 patients approached, 19 were enrolled in this study, resulting in a recruitment rate of 95% (figure 2). Baseline characteristics are presented in table 1.

Table 1

Patient characteristics and perioperative data

From the enrolled patients, two experienced significant postoperative bleeding and were not randomized. Nine patients were randomized to the block group and eight patients to the standard care group. One patient in the block group did not receive the intervention due to the presence of a vacuum dressing which obstructed the injection site. Thus, the adherence rate to randomization group was 94% (16/17).

No major adverse events were noted up to 48 hours of follow-up. Importantly, no pneumothorax or LA toxicity was reported (table 2).

Table 2

Feasibility data (primary outcomes)

The mean (SD) NRS pain scores at rest were 3.3 (3.2) in the block group vs 5.6 (3.2) in the control group at 12 hours (figure 3). With deep breathing, the pain scores were 4.0 (3.2) vs 7.8 (2.3) in the block and control group, respectively. At 24 hours, the mean NRS score was 4.1 (3.9) in the block group and 4.1 (3.3) in the standard care group. With deep breathing, the block and standard care groups were also similar, 5.4 (3.7) and 5.5 (3.9), respectively.

Figure 3

NRS pain scores at 12 and 24 hours. Data expressed as mean and SD. ICU, intensive care unit; NRS, Numeric Rating Scale.

The mean (SD) 24-hour hydromorphone requirement was 1.9 (1.1) mg in the block group vs 1.8 (0.9) mg in the control group. The acetaminophen administration at 24 hours was 1869 (810) mg in the block group and 1625 (919) mg in the control group. All eight patients who received the TTP block reported satisfaction with postoperative pain management at the end of the study, and seven of the eight patients in the standard care group were satisfied. Additional secondary outcomes are summarized in table 3.

Table 3

Secondary outcomes


Our pilot study revealed that performing a randomized controlled trial studying the TTP block in cardiac surgery patients is feasible and has a high recruitment rate. Given that this block is minimally invasive, single-shot, and ultrasound-guided, it is not surprising this block technique had high patient acceptability.

The adherence rate for this block was also high. The block requires minimal set-up, is relatively easy and quick to perform, and is performed on an already-sedated patient at a time of low patient care activities. As there were no indwelling catheter or pumps for the nurses to run, in addition to the possible benefit of fewer nursing interventions for pain, there were minimal workflow barriers to performing this block.

In the current study, we successfully demonstrated preliminary data indicating potential safety and feasibility of future clinical trials. We achieved high compliance with the study protocol from care providers, and there were no adverse events in our patient population attributable to the TTP block. Importantly, no pneumothorax or LA toxicity was reported in relation to this block. This is in keeping with our expectations for this superficial block, which is performed with a small-caliber needle at a site of low vascularity with a low dose of LA. Our findings are consistent with a recently published case series of 299 consecutive TTP block cases that noted only 2 incidences of ‘slight infections’ around the injection site.15 Two patients had postoperative bleeding and were excluded from randomization because they required close monitoring for hemodynamics, and the priority was given to stabilize the patients rather than to carry on with randomization.

As this was a feasibility study, we did not inferentially assess the statistical significance of pain scores, and this decision is supported by the extant literature on performing pilot studies.13 At 12 hours, our data suggest the block group experienced lower pain scores at rest and with deep breathing. At 24 hours the pain scores appear similar in both groups. This is consistent with the projected duration of 8–12 hours for most single-shot peripheral nerve blocks. The pain scores our patients reported at rest and with deep breathing are also consistent with previously reported levels of poststernotomy pain.1 3 Further study is warranted to assess the analgesic efficacy at these time points.

The 24-hour opioid requirements were similar in both groups. A possible explanation for this includes the hospital practice for opioid administration. Occasionally, opioids are given in the ICU to treat hypertension, to sedate patients, and to provide comfort for non-surgical musculoskeletal discomfort from immobility. Opioids are also used to treat chest tube and saphenous vein harvest site pain. We did not track the reasons for opioid administration, and more information may help us elucidate how much relief the TTP block provides for sternotomy pain.

The TTP block was widely accepted by the block group patients. The block did not cause any complications, such as bleeding from the injection site or pneumothorax, and did not interfere with surgical recovery or nursing care.

There are a few limitations to our study. As pain score was not our primary outcome, and we did not control for intraoperative opioid use and this may have affected postoperative pain scores; however, a review of intraoperative data across patients did not reveal large differences in intraoperative opioid management between groups. We also did not control for postoperative opioid use or document the reasons for opioid administration; therefore, the total 24-hour hydromorphone consumption may not be an accurate surrogate for postoperative pain. All blocks in this study were performed by one operator to ensure consistency; however, this reduces the external validity. In future trials, an increased number of operators will be necessary to determine if there is operator dependence as to the success of the TTP block.

Another limitation of our study is the lack of subgrouping of coronary artery bypass grafting (CABG) from other cardiac surgeries. Our group recently published a cadaveric TTP block study of a CABG patient who had an internal thoracic artery (ITA) conduit. We found the TTP block plane was surgically disrupted by ITA harvesting because the ITA courses in this plane.16 As a result, the TTP block injectate did not spread evenly between the desired thoracic levels. In this feasibility study, 60% of the patients underwent CABG with ITA harvesting. These patients may not have derived any benefit from the TTP block on the side of the ITA conduit. Patients undergoing CABG also experience pain from saphenous vein and radial artery harvest sites and pleural chest tubes, which the TTP block cannot treat.

A potential limitation of this block technique is the visualization of the transversus thoracis muscle by ultrasound. Since this muscle layer is thin and adjacent to the pleura, there is a risk of pleural puncture. Recently, Ohgoshi et al 17 and Del Buono et al 18 described the parasternal intercostal nerve block as a safer alternative to the TTP block, as the plane to be blocked is more superficial. Although this block has not been clinically used yet, it may be an effective alternative for patients with a history of ITA harvest.

In conclusion, the TTP block is a novel pain management strategy that shows promise poststernotomy. This pilot study established the feasibility of studying the TTP block in future prospective clinical trials to further examine its safety and analgesic efficacy. Our results reveal a high patient recruitment, adherence, and satisfaction rate, and provide some preliminary data supporting technique safety. More data comparing important clinical outcomes such as pain score, patient satisfaction, and potential adverse effects of postoperative opioid use, including the incidence of nausea and vomiting, respiratory complications, and prolonged postoperative mechanical ventilation, are needed.


The authors thank Rob Mayer, research assistant, for help with data collection. They also thank the entire staff of the Cardiac Surgery Recovery Unit at the London Health Sciences Centre for supporting this study.



  • Presented at The authors acknowledge that this study was presented as an oral presentation abstract at the 2018 Canadian Anesthesiologists Society Annual Meeting on June 17, 2018.

  • Contributors SF: study design, ethics application, patient recruitment, and manuscript preparation. MR: patient recruitment and manuscript preparation. PMJ: study design and manuscript preparation. DV: study design and manuscript preparation. DB: manuscript preparation. JRZ: study design, ethics application, patient recruitment, and manuscript preparation.

  • Funding Department of Anesthesia internal funding.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The study protocol was approved by the Research Ethics Board at Western University (REB ID#109015).

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles