Prevention of Postoperative Pain After Thyroid Surgery: A Double-Blind Randomized Study of Bilateral Superficial Cervical Plexus Blocks

Abstract

Local anesthetic infiltration may reduce postthyroidectomy pain. We performed a double-blinded, randomized, placebo-controlled trial to assess the analgesic efficacy of bilateral superficial cervical plexus blocks performed at the end of surgery. Ninety patients undergoing elective thyroid surgery by the same surgeon under general anesthesia were randomized to receive 20 mL isotonic sodium chloride or 20 mL bupivacaine 0.25% with 1:200,000 epinephrine. Postoperative pain was assessed every 4 h using an 11-point numeric rating scale (NRS-11). All patients received acetaminophen every 6 h. In addition, morphine was administered following a standardized protocol if the NRS-11 score was ≥4. The main outcome variables were pain scores (NRS-11), the proportion of patients given morphine at any time during the 24-h period, and the amount of morphine administered. The Bupivacaine group had a smaller proportion of patients given morphine (66.0% vs 90.0%;P = 0.016), and lower initial median pain scores (P = 0.002). We conclude that bilateral superficial cervical plexus blocks significantly reduce pain intensity in the postoperative period after thyroid surgery but do not provide optimal pain relief alone.

Many patients require acute pain control during the first day after thyroid surgery performed under general anesthesia. Few studies have addressed this issue ^(1,2). In one study, the mean pain score reported after thyroidectomy was 6.9 on a Visual Analog Scale (VAS) from 0 to 10 and 90% of patients required morphine ⁽¹⁾. In our center, postthyroidectomy pain intensity assessment in the postanesthesia care unit (PACU) shows that, despite acetaminophen administration, 70% of patients initially rate a score ≥4 on an 11-point numeric rating scale (NRS-11) (0 = no pain, 10 = worst imaginable pain). In this common short-stay type of procedure, gains can be made in improving pain control, a major component of quality of care.

Postthyroidectomy pain may be treated with nonsteroidal antiinflammatory drugs (NSAIDs) or opioids. However, many surgeons are reluctant to use NSAIDs so soon after this type of procedure because of fear of bleeding complications. Opioids have many well-known undesirable effects, including postoperative nausea and vomiting, which are frequent after this type of procedure ⁽³⁾.

Local anesthetic wound infiltration may reduce postthyroidectomy pain. Gozal et al. ⁽¹⁾ reported markedly reduced opioid requirements. The trial was randomized, but unfortunately the “blinding” was questionable because the caregivers were aware of the treatment. In another study, bupivacaine wound infiltration effectiveness was judged disappointing when compared with two opioid regimens ⁽²⁾. The goal of the present study was to assess in a double-blind fashion the analgesic efficacy of bilateral superficial cervical plexus blocks (BSCBs) performed with 0.25% bupivacaine with 1:200,000 epinephrine versus placebo, with regard to pain intensity scores and 24-h postoperative morphine requirements after thyroidectomy.

Methods

The study was approved by our institutional committee for human investigations and written informed consent was received from each patient. ASA physical status I-III adults scheduled for elective thyroid surgery under general anesthesia were eligible for the study. Patients were excluded if they had recently used opioids, corticosteroids, or any analgesic drug, or if they were unable to use the numeric scale scoring system for pain assessment. In addition, any patient with a history of intolerance to the medications used in the study was excluded.

All patients were instructed preoperatively about the use of NRS-11. Patients were premedicated with midazolam (0.1 mg/kg PO). Anesthesia was induced with sufentanil (0.25 μg/kg IV) and propofol (2–3 mg/kg IV). Orotracheal intubation was facilitated by the administration of atracurium (0.5 mg/kg IV). Anesthesia was maintained with isoflurane (0.5–1.2%) in a N₂O/O₂ (2:1) mixture. Supplemental sufentanil was given in incremental doses of 5–10 μg when necessary. Thyroid surgery was performed according to a standardized procedure by the same surgeon. Thirty minutes before the end of surgery, patients were administered propacetamol (2 g IV), a prodrug of acetaminophen.

Using a random number sequence, patients were randomized to receive either 20 mL isotonic sodium chloride solution (Saline group) or 20 mL bupivacaine 0.25% with 1:200,000 epinephrine (Bupivacaine group) as BSCBs after skin closure. BSCBs were performed as follows by the surgeon: in both sides of the neck, a subcutaneous infiltration of 10 mL of the solution was made using a 22-gauge needle, inserted at the midpoint of the sternocleidomastoid muscle, corresponding to the C3 transverse apophysis, in three directions. Six mL of the solution was injected up and down at the posterior border of the sternocleidomastoid muscle to block the occipital, auricular, and supraclavicular branches of the superficial cervical plexus and 4 mL was injected horizontally above the muscle to block the transverse cervical nerve. The injection was given using a sterile unlabelled syringe prepared outside the operating room by a staff member not involved in the patient’s care or pain assessment. The specific treatment given was unknown to the patient, the anesthesiologist, the surgeon, and the nurse in charge of pain assessment and management. No testing was done at any time to document that superficial cervical blocks were effective.

Patients were transferred to the PACU where they stayed for 2 h, then were transferred to the ward. At admission to the PACU, they were interviewed by a nurse trained in pain assessment when fully awake and extubated (H0), before discharge from the PACU (H+2), and every 4 h until 24 h after the end of surgery. Pain intensity was evaluated by the NRS-11 and by a four-category verbal rating scale (VRS-4) (no pain, mild pain, moderate pain, severe pain).

Postoperative pain management was standardized as follows. During the postoperative period, propacetamol (2 g) injections were repeated every 6 h during the first 24 h. In PACU, 2 mg morphine IV was repeated every 5 min until the NRS-11 score was lower than 4; if IV morphine titration reached 6 mg or more, the patient received 5 mg morphine subcutaneously before leaving the PACU. On the ward, the patient’s NRS-11 score was evaluated every 4 h by a nurse; if the pain score was 4 or higher, 5 mg morphine was administered subcutaneously. Prophylactic antiemetics were not administered.

The main outcome was the proportion of patients having a pain score ≥4 on the NRS-11 and subsequently given morphine at any time during the 24-h postoperative period. Other outcomes included pain scores on NRS-11 and pain categories on VRS-4 at admission to the PACU (H0), before discharge from the PACU (H+2), and at 6 h (H+6) and 24 h (H+24) after the end of surgery, and the amount of morphine required during the 24-h postoperative period. Nausea, vomiting, antiemetic requirements, sedation, and respiratory scores and other adverse events were fully recorded.

Assuming that BSCB would be successful in doubling the 30% proportion of patients not requiring morphine after thyroidectomy with α = 0.05 and β = 0.1, 90 patients were required in the study. Continuous variables are reported as mean ± sd and are analyzed using Student’s t-tests. Categorical variables are reported as proportions and were analyzed using χ² with continuity correction if appropriate. Pain scores and morphine consumption are reported as median and range because their distributions were expected to be skewed, and differences between groups were analyzed by using the nonparametric Mann-Whitney U-test. Statistical inference was evaluated at the 5% level with Bonferroni correction for multiple comparisons.

Results

Of the 90 patients enrolled in the study, 42 were in the Saline group and 48 in the Bupivacaine group. Three patients were excluded, two because of early reoperation for bleeding (one in each group) and one for protocol violation (Saline group). Demographic and clinical characteristics were similar in both groups (Table 1). An euthyroid state was obtained in all patients preoperatively. A high-vacuum wound drainage system (Drainobag^®Lock 150V; B. Braun, Melsungen AG, Germany) with Charriere 8 connecting tubing was used in all but two patients (one in each group).

Table 1: Demographic Characteristics and Operative Data

The proportion of patients with an NRS-11 score ≥4 at any time was significantly smaller in the Bupivacaine group, especially during the PACU period (Table 2). The proportion of patients reporting no/mild pain and moderate/severe pain at any time on the categorical VRS-4 was respectively 32% and 68% in the Bupivacaine group versus 8% and 93% in the Saline group (P = 0.011). Patients given bupivacaine had significantly lower NRS-11 scores on admission to the PACU (Fig. 1). NRS-11 scores were low and did not differ at any later postsurgical time studied, although the proportion of patients with a score ≥4 at H6 was 11% in the Bupivacaine group and 31% in the Saline group. Morphine cumulative doses were lower in the Bupivacaine group (Table 2).

Table 2: Morphine Requirements in Postanesthesia Care Unit (PACU) and During the First 24 Hours After Thyroidectomy

Figure 1: Numeric rating scale pain scores at admission to the postanesthesia care unit (H0), before discharge from the postanesthesia care unit (H+2), and at 6 h (H+6) and 24 h (H+24) after thyroidectomy. The box represents the 25th and 75th percentiles, and the median is represented by the solid line. Error bars above and below the box mark the 10th and 90th percentiles. * P = 0.0015.

Twenty patients experienced postoperative vomiting or retching (9 of 47 in the Bupivacaine group and 11 of 40 in the Saline group). No patient had significant sedation or respiratory depression requiring a prolonged observation period in the PACU.

Discussion

Postthyroidectomy pain perception likely includes many components linked to the deep and superficial layers of the wound, intraoperative neck position and wound drainage. The incisional infiltration of local anesthetics has been used for postoperative pain relief in different types of surgery, the best results being observed in superficial and short-lasting procedures such as herniorrhaphy ⁽⁴⁾. The expected decrease in pain scores or supplemental opioids has been disappointing in procedures with a major visceral component, such as major gynecological or cervical spine surgery ^(5,6). The relatively short duration of action of local anesthetics is frequently seen as a major limitation to their use for postoperative pain relief ⁽⁴⁾. The present study shows that BSCBs performed with 0.25% bupivacaine reduce postoperative pain intensity and opioid demand, supporting the view that postthyroidectomy pain has a large “superficial” component.

We chose to use the NRS-11 to measure pain intensity, although the VAS is seen as more sensitive. In a previous evaluation comparing the suitability of several pain intensity scales in this type of surgery, we found VAS unusable in approximately 25% of patients in the PACU, whereas NRS-11 was convenient in the early postoperative period. A study of the usefulness of the VAS in patients with acute postoperative pain found that it correlates well with the verbal 11-point scale ⁽⁷⁾. Simulation experiments have shown similar sensitivity of the NRS-11 and VAS when comparing acute postoperative pain intensity ⁽⁸⁾.

We found a larger than postulated proportion of patients given morphine in the Saline (control) group. Despite systematic acetaminophen administration in our patients, the proportion of morphine recipients in the Saline (control) group is the same (90%) as the one observed in the control group in the study by Gozal et al. ⁽¹⁾. Our expectation was based on previous local evaluations of postthyroidectomy pain management in the PACU. It may be that the larger than anticipated proportion of morphine recipients in the control group is the result of continuing improvements in postoperative pain care. Nevertheless, we cannot totally exclude that the 0.9% NaCl solution used as placebo worsened wound pain intensity. This method was deemed necessary to ensure that blinding would not be compromised, but introduced a potential bias. Other authors have already raised concerns about the advantages and limitations of saline as a placebo solution use in pain studies, and no study is able to confirm that saline infiltration is devoid of consequences on nociception ⁽⁹⁾.

Three cases were lost to the study, but this deletion had no impact on the outcome. The opioid analgesic sparing effect of BSCB was nonetheless weaker than expected from the data reported by Gozal et al. ⁽¹⁾ who used a similar technique. The authors infiltrated the cervical incision with 10 mL bupivacaine 0.5% and found that only 30% of patients required morphine when the wound was infiltrated. Differences in study design as well as in pain assessment and management produce differences in results. We cannot exclude the possibilities that a local wound infiltration of plain 0.5% bupivacaine is more effective than a bilateral injection of 0.25% bupivacaine with epinephrine near the emergence of the superficial cervical plexus branches. However, in the current trial, much attention was devoted to the blinding of the caregivers to the treatment, and the placebo saline infiltration was an important part of it.

The use of a BSCB cannot preclude the postoperative need for other analgesics in approximately 65% of patients. Whether better results could have been obtained with changes in the technique is debatable. The analgesic efficacy might have been improved with larger bupivacaine doses, but we were concerned about peak plasma bupivacaine concentrations that might occasionally result from the injection of such doses in a highly vascularized area. It is unlikely that preoperative administration of local anesthetics would have been more effective than postoperative administration. Although many studies have suggested that the timing of administration of analgesic drugs could influence their efficacy by reducing the sensitization of the nervous system induced by the nociceptive inputs, this concept of preemptive analgesia is debated in both clinical and basic science research ⁽¹⁰⁾. The residual pain probably results from the contribution of nociceptive stimuli arising from areas that cannot be blocked by a superficial approach. Further studies are required to evaluate this assumption because the blocks were not evaluated in the current study to keep the assessors blind to the group allocation. The combination of superficial and deep cervical plexus blocks might have provided lower pain scores and morphine requirements. Such a combination allows surgery in the anterolateral cervical area and has even been used for thyroid surgery ^(11,12). However, bilateral deep cervical plexus blocks can produce a highly dangerous bilateral diaphragmatic dysfunction. In view of the potential risks, we consider that this potentially unsafe technique should not be used for postoperative pain relief.

The main benefit of BSCB to the patient was lower pain intensity scores in the early postoperative period. The reduced opioid consumption was less clinically relevant, as it did not offer any clear benefit to the patient in terms of reduced side effects. Emesis is dependent on a variety of factors and the study design did not cover this issue. Expecting an impact on such an outcome was beyond the scope of the present study. Clinically significant sedation and respiratory depression did not occur, even in patients with the largest morphine doses. Although patient satisfaction was not evaluated in the study, improving postoperative pain prevention is in itself an important outcome from the patient’s view ⁽¹³⁾.

In summary, a BSCB significantly reduces pain intensity scores in the early postoperative period after thyroid surgery, but optimal pain relief cannot be achieved by a BSCB alone. A BSCB may be integrated into a multimodal approach ⁽¹⁴⁾.