Hybrid Parasternal Intercostal Plane Block Versus No Block for Postoperative Analgesia After Cardiac Surgery Via Median Sternotomy

BACKGROUND Pain after median sternotomy is multifactorial, arising from the sternotomy itself, costochondral injury, sternal retraction, chest tubes, and graft harvest sites. Poorly controlled pain contributes to nausea, vomiting, delirium, arrhythmia, and myocardial ischemia, and delays recovery. Enhanced Recovery After Cardiac Surgery (ERAS Cardiac) guidelines recommend regional anesthesia as part of opioid-sparing multimodal analgesia. The parasternal intercostal plane (PIP) block, which targets the anterior cutaneous branches of the T2-T6 intercostal nerves, has emerged as a practical option in cardiac surgical patients who receive high-dose anticoagulation, in whom neuraxial or paravertebral techniques carry greater risk.

PIP blocks are classified by injection depth. The superficial PIP block (S-PIP) deposits local anesthetic between the pectoralis major and internal intercostal muscles, whereas the deep PIP block (D-PIP) deposits it between the internal intercostal and transversus thoracis muscles. Both techniques have been shown to reduce postoperative pain and opioid consumption compared with no block. However, D-PIP is performed closer to the pleura and internal mammary artery (IMA). In a recent three-arm randomized trial (n = 254), the incidence of pneumothorax was higher with D-PIP than with S-PIP or control (11.8% vs 4.8% vs 2.4%, p = 0.032). On the left side, the risks of D-PIP are amplified when the left IMA is harvested for coronary artery bypass grafting, and the left-sided deep approach is ergonomically more demanding for right-handed operators. Conversely, abandoning D-PIP entirely may sacrifice its potential advantages, including wider dermatomal coverage (including T1) and better cranio-caudal spread from a single injection.

RATIONALE The hybrid PIP block (HPIP) allocates the two techniques asymmetrically: S-PIP on the left side, where procedural risk and technical difficulty are concentrated, and D-PIP on the right side, where these constraints are fewer. This strategy (1) limits exposure to pneumothorax and IMA injury to one side, (2) avoids the side of left IMA harvest so that it can be applied to all sternotomy cardiac surgeries including coronary bypass grafting, (3) avoids the technically demanding left-sided deep approach, and (4) may partly preserve the upper-dermatome and midline sternal analgesia provided by the deep plane. Because no study has evaluated this asymmetric combination, its efficacy cannot be inferred from studies of either technique alone. This trial is the first step to establish whether HPIP provides clinical benefit over standard multimodal analgesia without a block.

STUDY DESIGN This is a prospective, randomized, patient- and assessor-blinded, single-center, parallel-group superiority trial conducted at Asan Medical Center, Seoul, Republic of Korea. A total of 172 adults scheduled for elective cardiac surgery via median sternotomy will be randomized 1:1 to the HPIP group or the No-block group.

RANDOMIZATION AND BLINDING An independent investigator will generate the allocation sequence by computer using block randomization with variable block sizes of 2-6. Allocation will be concealed in sequentially numbered, opaque, sealed envelopes opened on the day of surgery. The anesthesiologist performing the block cannot be blinded but will not participate in postoperative pain assessment, data collection, or rescue analgesia decisions. Patients, the surgical team, ward and ICU nurses, outcome assessors, and the statistician will remain blinded. Because the block is performed under general anesthesia and leaves no visible mark, no sham procedure will be performed in the control group, to avoid exposing participants to unnecessary invasive risk. Block performance will be recorded only in research-specific documents, not in shared medical or anesthesia records.

INTERVENTIONS HPIP group: After induction of general anesthesia and before skin incision, an experienced anesthesiologist performs an ultrasound-guided in-plane block approximately 2 cm lateral to the sternum. On the left side, S-PIP is performed with 10 mL injected at each of the 2nd and 4th intercostal spaces. On the right side, D-PIP is performed with a single 20 mL injection at the 3rd intercostal space. Each side receives 20 mL of 0.375% ropivacaine (75 mg) mixed with dexamethasone 2.5 mg, for a total of 40 mL (ropivacaine 150 mg, dexamethasone 5 mg). Ropivacaine is reduced to 0.25% (100 mg in total) in patients with ideal body weight < 50 kg, Child-Pugh B hepatic impairment, or eGFR 15-29 mL/min/1.73 m².

No-block group: No chest wall fascial plane block is performed.

Both groups receive identical general anesthesia and postoperative multimodal analgesia: scheduled oral acetaminophen (switched to acetaminophen/ibuprofen/codeine if needed), fentanyl-based IV patient-controlled analgesia (basal 1 mL/h, bolus 1 mL, lockout 15 min), and stepwise rescue analgesia (tramadol, fentanyl, pethidine, then ketorolac) for NRS ≥ 4 or CNPS ≥ 3. All opioid doses are converted to morphine milligram equivalents (MME).

SAFETY MONITORING In the HPIP group, a standardized lung ultrasound protocol is applied to screen for block-related pneumothorax. Lung sliding is assessed bilaterally before the block, as part of the routine anterior chest ultrasound performed after central venous catheterization, and again immediately after the block and before skin incision. If lung sliding is newly absent, a lung point is sought and chest radiography is performed for confirmation. Postoperative adverse events, including pneumothorax, hemothorax, hematoma, nerve injury, injection-site infection, local anesthetic systemic toxicity, PONV, and urinary retention, are monitored at 0, 6, 12, 24, and 48 hours and until hospital discharge.

SAMPLE SIZE Based on a meta-analysis of PIP block versus control (mean difference in 12-hour pain score -1.21 on a 0-10 scale) and a weighted pooled standard deviation of 2.6 back-calculated from the included studies, 73 evaluable patients per group are required to detect a difference of 1.21 points with a two-sided alpha of 0.05 and 80% power. Allowing for a 15% dropout rate, 86 patients per group (172 in total) will be enrolled.

STATISTICAL ANALYSIS The primary analysis will be performed in the modified intention-to-treat population (all randomized patients with at least one primary outcome assessment), with a per-protocol analysis as a sensitivity analysis. The primary outcome will be analyzed using analysis of covariance adjusted for baseline characteristics, reporting the mean difference (HPIP minus No-block) with a two-sided 95% confidence interval. Superiority will be concluded if the upper bound of the confidence interval is below zero. Prespecified sensitivity analyses will address the effect of sedation on NRS assessment. Missing data will be handled by multiple imputation, with best-worst and worst-best scenario analyses and, if needed, a tipping-point analysis. All secondary and safety outcomes are exploratory and will not be adjusted for multiplicity.