The purpose of this study was to investigate whether continuous paravertebral block at levels T1 and T2 with bupivacaine infusion can improve the survival of free flaps in maxillofacial reconstruction. The study was designed as a randomized controlled trial and included 36 adult patients scheduled for maxillofacial free flap reconstruction under general anesthesia. Patients were randomly divided into two groups: patients in group A received continuous paravertebral block at levels T1 and T2, while patients in group B served as controls. Postoperatively, a skin thermometer was used to assess the skin temperature. Perfusion of the flaps was evaluated by analysis of skin color, turgor, and capillary refill. Survival of the free flap was recorded. The surface temperature of the reconstructive flap, skin color score, and capillary refill score were significantly higher in group A patients than in group B patients during follow-up. The total perfusion score was significantly higher in group A than in group B at 16 h and 20 h postoperative ( P = 0.041 and P = 0.039, respectively). Re-operation was recorded in three cases in group B (16.7%) ( P = 0.031). Continuous paravertebral block at levels T1 and T2 can increase the skin temperature and improve skin color and capillary refilling, which are indices of adequate tissue perfusion and indicate maxillofacial free flap survival.
Anesthesia can be an important factor in improving the success rate of microvascular free flap surgery through the control of global hemodynamics and regional blood flow . Regional anesthesia at the recipient site has been associated with a better outcome and fewer flap complications . The continuous provision of this block after surgery might also be beneficial in preventing postoperative vasospasm in the reconstructed tissue .
The sympathetic innervation to the head and neck originate in the first and second thoracic spinal cord segments (T1 and T2). Paravertebral block at levels T1 and T2 blocks the preganglionic sympathetic fibers designated for the head and neck before they ascend to contribute in the cervical sympathetic ganglia . The present study aimed to investigate whether continuous paravertebral block at T1 and T2 with bupivacaine infusion can improve the survival of free flaps used for reconstruction in the maxillofacial area. The study sought to determine whether continuous paravertebral block at T1 and T2 improves the flap tissue temperature as the primary outcome measure; other parameters related to tissue perfusion were assessed as secondary outcomes.
Materials and methods
This study included 36 adult patients treated in the study department between 2012 and 2015. These patients had maxillofacial defects, were American Society of Anesthesiologists (ASA) physical status 2 or 3, and were scheduled for maxillofacial free flap microvascular surgery under general anesthesia. The study was designed as a randomized controlled trial (randomization was done by closed envelope technique). Patients were randomly divided into two equal groups: patients in group A received continuous paravertebral block at levels T1 and T2, while patients in group B did not receive the paravertebral block and served as controls. Group B patients did not receive a placebo of normal saline to avoid the potential risks of infection or hematoma.
Inclusion criteria were (1) patients with extraoral maxillofacial defects requiring surgery and reconstruction using free flaps; (2) ASA physical status 2 or 3.
The following exclusion criteria were applied: (1) patients for whom general anesthesia was contraindicated due to severe systemic medical problems (ASA physical class 4–6); (2) patients with bleeding diathesis (abnormal prothrombin time, partial thromboplastin time, or platelet count); (3) patients with hypersensitivity to local anesthetics; (4) patients with cervical spine disorders; (5) patients with intraoral defects, to exclude the thermal effect of the inside of the mouth when recording the flap skin paddle temperature.
On arrival in the operating room, routine monitoring was set up using a multichannel monitor (Infinity Vista; Dräger, Lübeck, Germany) and a 16-gauge intravenous cannula was inserted.
Paravertebral block was performed for group A patients with full aseptic precautions. The patient was seated with the neck and back flexed. A single-level paravertebral block at or below the mid-dermatomal level was performed (T1 or T2 is an appropriate level). Skin infiltration with local anesthetic was required. A 21-gauge needle was then used to infiltrate down to the transverse process. A volume of 2 ml of 2% lidocaine was sufficient.
The superior border of the spinous process was identified; the site of introduction of the needle was 2.5 cm lateral from the spinous process. For single paravertebral block, a 22-gauge Tuohy needle (B-Braun Medical Inc., Bethlehem, PA, USA) was introduced perpendicularly to contact the posterior surface of the transverse process. When the transverse process was contacted, the distance between the skin and the transverse process was established; the needle was then withdrawn to the skin surface and reintroduced 1 cm beyond the transverse process at a 15° to 60° angle, allowing the positioning of the needle below the transverse process. It was possible to experience a loss of resistance when the needle was pushed beyond the transverse process through the costo-transverse ligament. After the paravertebral space had been located, 5–10 ml of 2% lidocaine was injected to expand the space and a standard epidural catheter was then advanced no more than 2 cm into the space .
Continuous paravertebral block was started with a loading dose of 10 ml 0.25% bupivacaine, followed by an infusion of 4–5 ml of 0.25% bupivacaine per hour for up to 48 h after surgery.
It was ensured that the blockade was working before the induction of general anesthesia. Patients with successful blockade will present Horner’s syndrome on the blockade side: miosis, ptosis, and anhydrosis. If the blockade was unsuccessful, the patient was excluded from the study.
All patients underwent the same anesthetic protocol: intravenous (IV) propofol 2–3 mg/kg and IV fentanyl 1.5 μg/kg; endotracheal intubation was facilitated by IV rocuronium 0.6 mg/kg. Anesthesia was maintained with 2–3% sevoflurane in oxygen, IV fentanyl, and rocuronium increments. In addition to standard monitoring, urinary output was recorded. Patients were not extubated in the operating room and were moved to the surgical care unit for further management.
Heart rate (HR) and mean arterial blood pressure (MAP) were measured preoperatively, intraoperatively (after induction and every 15 min), and postoperatively (every 15 min for 1 h and every 4 h for 48 h). The patients received standard postoperative analgesia with meperidine 0.5 mg/kg every 8 h for 48 h. The intensity of postoperative pain was assessed every 4 h for 48 h using a standard visual analog scale (VAS). Patients received IV analgesia according to their VAS score: for a score of 0–3, nothing was given; for a score of 4–6, diclofenac sodium 75 mg was given, with a maximum dose of 150 mg/day; for a score of >6, meperidine 0.5 mg/kg was given. The total postoperative analgesic requirement in the first 48 h postoperative was calculated.
A skin thermometer (Feverscan; LCR Hallcrest, Flintshire, UK) was used every 4 h to assess the skin temperature on the surgical side (Temp1) and on the contralateral non-surgical side (Temp2) in both study groups for 48 h after surgery. In addition, the perfusion of the flaps was assessed through the evaluation of skin color, turgor, and capillary refill; these were graded every 4 h according to the scores shown in Table 1 . Survival of the free flap was recorded in both groups.
|Skin color (1–4)||Pale/white||Blue/purple||Deep red||Pink|
|Capillary refill (1–4)||Absent||Sluggish >2 s||Fast <1 s||Brisk (1–2 s)|
The guidelines of the Declaration of Helsinki were followed in this investigation. The institutional review board (IRB) of the university hospital approved the study design. Written informed consent was obtained from every patient.
Data were recorded as the mean ± standard deviation values. Comparisons were made with the t -test. Data were analyzed using IBM SPSS Statistics version 20 software (IBM Corp., Armonk, NY, USA); a P -value of <0.05 was considered significant.
Sample size calculation
A sample size of 15 patients per group was required to achieve 90% power to detect a difference of 3.55 between (1) the null hypothesis that the mean difference in surface temperature between the reconstructed side and contralateral non-surgical side (ΔTemp = Temp1 − Temp2) in both groups is −3.02 and (2) the alternative hypothesis that the mean difference in the group receiving continuous block is 0.53, with estimated group standard deviations of 0.34 and 1.27 for groups A and B, respectively, and a significance level (alpha) of 0.01 using the two-sided independent sample t -test, assuming the actual distribution to be normal .
No patient was excluded from the study due to a failed block. No significant difference was found between the two patient groups with regard to age, sex, body weight, ASA physical status, or duration of surgery ( Table 2 ).
|Age (years)||45.6 ± 12.5||46.1 ± 11.34||0.58|
|Weight (kg)||65.34 ± 10.9||71.44 ± 10.5||0.107|
|Duration of surgery (h)||7.63 ± 1.87||7.91 ± 2.03||0.711|
Recipient vessels used for the free flaps were the facial artery and facial or internal jugular vein. The anastomoses were performed end-to-end with the facial vessels and end-to-side with the internal jugular vein. The free flaps employed in this study were the anterolateral thigh flap ( n = 18), radial forearm flap ( n = 10), and deep inferior epigastric artery flap ( n = 8); the flap types were distributed equally between the two groups.
No recurrence of the primary tumor occurred, and there were no patient deaths. There was no major complication as a result of the surgical procedure. The follow-up period ranged from 6 months to 2.5 years, with a mean follow-up period of 14.6 months.
There were no statistically significant differences in preoperative, intraoperative, or postoperative HR and MAP between the two study groups. For both groups, there was no significant change in intraoperative or postoperative HR or MAP as compared with the preoperative values. There was no statistically significant difference between the two study groups regarding intraoperative or postoperative urine output ( Table 3 ). There was also no statistically significant difference between the two study groups regarding the postoperative pain VAS or postoperative analgesic requirements ( Table 3 ).
|Intraoperative urine output (l)||1.67 ± 0.28||1.59 ± 0.24||0.521|
|Postoperative urine output (l)||3.81 ± 0.89||3.89 ± 0.92||0.975|
|Postoperative diclofenac sodium (mg/48 h)||280 ± 30.9||268.5 ± 20.3||0.36|
|Postoperative meperidine (mg/48 h)||85.9 ± 13.6||92.6 ± 11.6||0.104|
|Flap re-operation %||0 (0.0%)||3 (16.7%)||0.031 a|
The surface temperature of the reconstructive flap (Temp1) was higher in patients receiving continuous paravertebral block (group A) than in the patients without the block (group B), and this was statistically significant at 0, 12, 16, and 20 h postoperative ( P = 0.033, P = 0.021, P = 0.016, and P = 0.014, respectively ( Fig. 1 ). The maximum difference in Temp1 between the two groups was 2 °C, occurring at 16 h and 20 h after surgery. The difference in surface temperature between the reconstructive site and non-surgical site (ΔTemp = Temp1 − Temp2) was greater in group A than in group B ( Fig. 2 ). Statistically significant differences in ΔTemp were found at 8, 12, 16, and 20 h after surgery ( P = 0.049, P = 0.023, P = 0.006, and P = 0.006, respectively).
Reconstructive flap perfusion was assessed in terms of skin color, turgor, and capillary refill scores. These scores were higher in group A patients (receiving continuous paravertebral block) than in group B patients (without the block). Skin color and capillary refill scores were significantly higher in group A at 16 h and 20 h postoperative ( P = 0.031 and P = 0.031, respectively, for skin color; P = 0.013 and P = 0.004, respectively, for capillary refill), while the skin color score did not differ significantly between the groups at any time ( Tables 4–6 ).
|Follow-up, h||Group A
|0||3.27 ± 0.46||3.20 ± 0.41||0.458|
|4||3.33 ± 0.49||3.27 ± 0.46||0.512|
|8||3.40 ± 0.51||3.27 ± 0.46||0.107|
|12||3.47 ± 0.52||3.27 ± 0.46||0.098|
|16||3.60 ± 0.51||2.87 ± 0.52||0.013 b|
|20||3.53 ± 0.52||2.40 ± 0.54||0.004 b|
|24||3.53 ± 0.52||3.33 ± 0.49||0.101|
|28||3.60 ± 0.51||3.40 ± 0.51||0.118|
|32||3.53 ± 0.52||3.33 ± 0.49||0.101|
|36||3.53 ± 0.52||3.33 ± 0.49||0.101|
|40||3.53 ± 0.52||3.33 ± 0.49||0.101|
|44||3.60 ± 0.51||3.40 ± 0.51||0.118|
|48||3.53 ± 0.52||3.53 ± 0.52||1.000|