Does Surgical-site Multimodal Drug Injection After Palmar Plating of Distal Radius Fractures Improve Pain Scores?

Abstract

Background

Although palmar locked plating is a stable fixation method frequently used to treat unstable distal radius fractures (DRFs), surgical treatment may be painful, and so interventions to decrease that pain might improve our patients’ experiences with surgery. Some surgeons use local multimodal drug injections to decrease postoperative pain after lower-extremity arthroplasty, but little is known about the effectiveness of a local multimodal drug injection in patients who undergo palmar plating for DRFs.

Questions/purposes

(1) Do patients who receive a local multimodal drug injection after palmar plating for unstable DRFs have better pain scores at 4, 8, 24, and 48 hours after surgery than patients who have not received such an injection? (2) Do patients who receive a local multimodal drug injection have lower fentanyl consumption and administration of anti-emetic drugs within the first 48 hours after surgery than patients who have not received such an injection?

Methods

A randomized controlled study was performed between August 2018 and August 2019 at a single tertiary care referral center. Patients who underwent palmar plating for DRFs under general anesthesia were eligible for inclusion. Patients were allocated into two groups: Those who received a local multimodal drug injection, and those who did not receive an injection. During the study period, 101 patients treated with palmar plating for DRFs met the inclusion criteria and were enrolled and randomized. Fifty-two patients were allocated to the multimodal injection group and 49 were allocated to the control group. Three patients (two in the multimodal injection group and one in the control group) were excluded after randomization because their pain level was not registered at any timepoint and so they could not be analyzed; our analysis was by intention to treat, and there was no crossover. After palmar plating, patients in the multimodal injection group received an injection of ropivacaine (10 mL), morphine (5 mL), ceftezole (5 mL) as well as normal saline (5 mL) to the periosteal area, pronator quadratus muscle, subcutaneous area, and skin. There were no differences between the groups in terms of age (62 years ± 13 years in the multimodal injection group versus 62 years ± 11 years in the control group; p = 0.93), gender (84% [42 of 50] women in the multimodal injection group versus 77% [37 of 48] women in the control group; p = 0.39), hand dominance (70% [35 of 50] dominant wrist in the multimodal injection group versus 60% [29 of 48] dominant wrist in the control group; p = 0.32) and AO/Orthopaedic Trauma Association (AO/OTA) classification (p = 0.57). All patients underwent treatment with the same perioperative protocol, and 25 μg of fentanyl was injected intravenously when a patient complained of pain and asked for additional pain control after surgery. In addition, when a patient complained of nausea or vomiting associated with fentanyl use, an anti-emetic drug was also injected. All nursing staff who administered the analgesics and anti-emetic drugs were blinded to treatment allocation. These two groups were compared regarding their pain level using a 100-mm VAS at 4, 8, 24, and 48 hours postoperatively. The minimum clinically important difference (MCID) for the VAS score was set to 20 mm. VAS scores were also collected by nursing staff who remained blinded to the treatment allocation. The total amount of fentanyl use and the number of patients who received anti-emetic drugs associated with administration of fentanyl within the first 48 hours were also recorded.

Results

With an MCID of 20 points, we found no clinically important reduction in VAS scores among patients who received a local multimodal injection compared with those who did not receive an injection at 4 hours (34 ± 15 versus 41 ± 20, mean difference -7.079 [95% CI -13.986 to -0.173]; p = 0.045), 8 hours (27 ± 16 versus 40 ± 19, mean difference -12.263 [95% CI -19.174 to -5.353]; p = 0.001), 24 hours (18 ± 12 versus 29 ± 20, mean difference -11.042 [95% CI -17.664 to -4.419]; p = 0.001), and 48 hours (9 ± 8 versus 10 ± 6, mean difference -1.318 [95% CI -4.000 to 1.365]; p = 0.33). Within the first 48 hours after surgery, fentanyl consumption was lower in patients receiving a local multimodal injection than in control patients (25 μg [range 0-100 μg] versus 37.5 μg [range 0-125 μg], difference of medians -12.5; p = 0.01). There was also a difference between the study groups in terms of the proportion of patients who received anti-emetic medications (16% [8 of 50] in the multimodal injection group versus 35% [17 of 48] in the control group, odds ratio = 2.879 [95% CI 1.102 to 7.519]; p = 0.03).

Conclusions

Our data suggest that patients who received a surgical-site multimodal analgesic injection after palmar plating for a distal radius fracture had no clinically important reduction in pain scores, but they did consume lower doses of opioid analgesics and fewer of these patients received anti-emetic drugs within 2 days of surgery. The high-potency opioids or other analgesia usually used for postoperative pain management have many side effects. Thus, reducing additional analgesia is as important as postoperative pain management and a surgical-site multimodal analgesic injection is one of the methods to achieve this a goal.

Level of Evidence

Level I, therapeutic study.

Introduction

Palmar locked plating is frequently used to treat unstable distal radius fractures (DRFs) and is a stable fixation method for early ROM [8, 12, 24]. Various options may control postoperative pain, such as anti-inflammatory drugs, regional or peripheral nerve blocks, and patient-controlled analgesia such as opioids [7, 17, 27]. Opioids are powerful analgesics and one of the usual options for postoperative pain management. However, these agents have notable adverse effects including nausea, emesis, pruritus, constipation, ileus, urinary retention, respiratory depression, confusion, and risk of dependency [5, 16, 29]. To reduce opioid-related side effects and improve analgesia, multimodal pain management has been introduced. Multimodal pain management with preoperative medication and a surgical-site injection during surgery involves the use of multiple agents that act on different pain pathways and provide proper pain relief, with less dependence on opioids [21]. In particular, a surgical-site injection has been shown to control postoperative pain effectively and avoid many of the side effects associated with systemic narcotic use [1, 14, 15, 21, 26-28].

Although numerous previous studies have addressed postoperative pain management with several multimodal pain techniques after orthopaedic procedures [13, 15, 19, 21, 26, 30, 31], information on pain levels or pain control after palmar plating of DRFs is very limited. Postoperative pain after palmar plating might be underestimated by patients and physicians, because postoperative pain is considered unavoidable after fracture surgery. This contrasts with pain after elective surgery such as lower-extremity arthroplasty [9, 22]. Nonetheless, pain control should be improved because controlling pain and swelling in the early stages can restore maximal function with minimal complications in later rehabilitation stages [6, 18].

Therefore, we asked: (1) Do patients who receive a local multimodal drug injection after palmar plating for unstable DRFs have better pain scores at 4, 8, 24, and 48 hours after surgery than patients who have not received such an injection? (2) Do patients who receive a local multimodal drug injection have lower fentanyl consumption and administration of anti-emetic drugs within the first 48 hours after surgery than patients who have not received such an injection?

Patients and Methods

We performed a randomized controlled study at a single center between August 2018 and August 2019. The study was registered in the Clinical Research Information Service, which is a primary registry of the World Health Organization’s International Clinical Trials Registry Platform (trial number: KCT0003313). This study was also approved by the institutional review board at our hospital and all patients provided informed consent to participate. This study is reported as per the Consolidated Standards of Reporting Trials (CONSORT) guideline. We included patients with DRFs without an ulnar styloid fracture or with an ulnar styloid fracture that did not lead to surgery. The surgical indications (only one criterion was needed for the participating surgeons to recommend palmar plating) were radial shortening greater than 5 mm, dorsal angulation greater than 20°, radial inclination less than 15°, articular step-off greater than 2 mm, and dorsal comminution greater than 50% [23]. We excluded patients with multiple injuries, bilateral or open DRFs, neurovascular injuries, combined distal radioulnar joint instability, associated carpal instability or a large ulnar styloid fragment treated with fixation, psychiatric illness, a known allergy or contraindication to opioids or local anesthetics, and those who regularly used narcotics and underwent surgery after 2 weeks. Patients who wanted a brachial plexus blockade or patient-controlled analgesia postoperatively were also excluded.

Randomization was performed via Microsoft Excel (Microsoft, Redmond, WA, USA) to generate random numbers. Group allocations were made by a research associate (S-HB) who did not participate in the study’s recruitment process or outcomes assessments. Allocations were concealed in a password-protected database accessible only to the research associate and were revealed to the surgery team 1 hour before the procedure began. All included patients were also blinded to treatment allocation.

The patients were divided into two groups: the multimodal injection group, those who received intraoperative surgical-site injections, and the control group, those who did not receive injections. For the multimodal injection group, a 25-mL anesthetic cocktail of 75 mg of ropivacaine (10 mL), 5 mg of morphine sulphate (5 mL), 1000 mg of ceftezole (5 mL), and normal saline solution (5 mL) was prepared in 50-mL syringes. After palmar plating, 10 mL of the anesthetic cocktail was injected into the periosteal area of the fracture site and pronator quadratus muscle. After pronator quadratus repair, the remaining 15 mL of the anesthetic cocktail was injected into the subcutaneous area and skin. All patients underwent treatment with the same perioperative protocol. When surgery was scheduled, patients received 650 mg of acetaminophen three times and 50 mg of tramadol twice per day as a pre-emptive analgesic [9]. Pain medication taken by patients before they visited our clinic was not recorded. After surgery, all patients also received 50 mg of tramadol every 12 hours and 650 mg of acetaminophen every 8 hours as postoperative medications for 7 days. To provide more appropriate postoperative pain management, patients and nurses were instructed to use fentanyl as an additional method of pain control. Thus, when a patient complained of pain and asked for additional pain control, duty nurses intravenously injected 25 μg of fentanyl. In addition, when a patient complained of nausea or vomiting associated with fentanyl use, nursing staff injected an anti-emetic drug (metoclopramide).

Of the 142 patients eligible for the study, 101 were enrolled and randomized. Fifty-two patients were allocated to the multimodal injection group and 49 were allocated to the control group. Three patients (two in the multimodal injection group and one in the control group) were excluded after randomization because their pain level was not registered at any timepoint (Fig. 1) and so they could not be analyzed; our analysis was otherwise by intention to treat, and there was no crossover. The mean age of the overall sample was 62 years ± 12 years, and 81% were women. AO/OTA C1 fractures were the most common fracture (49%) in both groups. The mean time from injury to surgery was 5 days ± 3 days in the overall sample, with no differences between the groups. There were also no differences between the groups in any of the demographic characteristics including age, sex, BMI, hand dominance, smoking status, and AO/OTA classification (Table 1).

Fig. 1.

This Consolidated Standards of Reporting Trials flowchart shows the trial enrollment and analysis.

Table 1.

Demographic and fracture characteristics of the groups

Characteristics	Multimodal injection group (n = 50)	Control group (n = 48)	p value
Age (years)	62 ± 13	62 ± 11	0.93
BMI (kg/m2)	25 ± 4	25 ± 4	0.99
Percentage of patients who were women	84 (42)	77 (37)	0.39
Percentage of patients with injury to the dominant wrist	70 (35)	60 (29)	0.32
Percentage of patients who smoke	6 (3)	8 (4)	0.48
Time from injury to surgery (days)	5 ± 2	4 ± 3	0.39
AO/OTA classification			0.57
Percentage of A2	12 (6)	21 (10)
Percentage of A3	16 (8)	10 (5)
Percentage of B3	6 (3)	4 (2)
Percentage of C1	46 (23)	52 (25)
Percentage of C2	20 (10)	13 (6)

Values are presented as the mean ± SD or % (n); AO/OTA = AO/Orthopaedic Trauma Association.

Surgical Procedure

All patients underwent surgery under general anesthesia. A small skin incision measuring approximately 2.5 cm was made, and the fracture was exposed using a palmar Henry approach. The pronator quadratus was completely detached from the radial insertion. Open reduction was performed with direct visualization of the fracture fragments, and the palmar locking plate was fixed under fluoroscopic control. Only one type of palmar radius locking plate (DVR, Biomet, Miami, FL, USA) was used during the study. A silicone drain was used routinely to minimize hematoma formation. One experienced hand surgeon (JSL) performed all surgeries during the study period. After surgery, wrists were immobilized using a short-arm palmar plaster splint, and patients were encouraged to perform active shoulder, elbow, and finger motion exercises. At 2 weeks after surgery, all wrists were placed in removable wrist braces.

Outcome Measures

The primary outcome variable was postoperative pain levels. Patients were asked to rate pain using a 100-mm VAS 1 hour before surgery as well as at 4, 8, 24, and 48 hours after surgery [25]. The minimum clinically important differences (MCID) for the VAS score was set to 20-mm [2]. Nursing staff who remained blinded to the treatment allocation collected VAS scores. The mean preoperative VAS score was 13 ± 6 in multimodal injection group and 11 ± 6 in control group (mean difference 2.098 [95% CI -0.259 to 4.454]; p = 0.08).

The secondary outcome variables were total fentanyl consumption during the first 48 hours after the surgical procedure, and the number of patients who received anti-emetic medications associated with fentanyl administration. All nursing staff who administered the analgesics and anti-emetic drugs were blinded to treatment allocation. Physician staff who knew of the intraoperative randomization process were not involved in recording any outcome measures.

Statistical Analysis

To estimate the required sample size, we conducted a pilot study by measuring change in the VAS score at 4 hours after surgery for 10 patients per group. The mean VAS score in multimodal injection group and control group were 35 ± 14 and 47 ± 21, respectively. With a two-tailed p value of 0.05 and power of 80%, we needed 36 patients in each group. Anticipating a loss-to-follow-up incidence of 20%, we set the target enrollment at 45 patients in each group. We used a t-test or Mann-Whitney U test to determine the significance of group differences in continuous variables, and we used a chi-square or Fisher’s exact test to determine the significance of differences between categorical variables. Statistical significance was defined as p < 0.05.

Results

With an MCID of 20 points, we found no clinically important reduction in VAS scores among patients who received a surgical-site multimodal injection compared with those who did not receive an injection at 4 hours (34 ± 15 versus 41 ± 20, mean difference -7.079 [95% CI -13.986 to -0.173]; p = 0.045), 8 hours (27 ± 16 versus 40 ± 19, mean difference -12.263 [95% CI -19.174 to -5.353]; p = 0.001), 24 hours (18 ± 12 versus 29 ± 20, mean difference -11.042 [95% CI -17.664 to -4.419]; p = 0.001) and 48 hours (9 ± 8 versus 10 ± 6, mean difference -1.318 [95% CI -4.000 to 1.365]; p = 0.33). The highest pain levels in both groups were experienced at 4 hours postoperatively and gradually diminished over time (Fig. 2).

Fig. 2.

VAS scores (expressed as the mean and standard error) of multimodal injection group and control group are shown.

Within the first 48 hours after surgery, fentanyl consumption was lower in patients receiving a local multimodal injection than in control patients (25 μg [range 0-100 μg] versus 37.5 μg [range 0-125 μg], difference of medians -12.5; p = 0.01) (Fig. 3). There was also a difference between the study groups in terms of the proportion of patients who received anti-emetic medications (16% [8 of 50] in the multimodal injection group versus 35% [17 of 48] in the control group, odds ratio = 2.879 [95% CI 1.102 to 7.519]; p = 0.03).

Fig. 3.

This boxplot shows the total amount of fentanyl consumption during the first 2 postoperative days.

Discussion

There has been increasing interest in multimodal management of postoperative pain, early rehabilitation, and reduction of opioid consumption after orthopaedic surgery. Nonetheless, many previous studies on multimodal management have been limited to the arthroplasty surgery [13, 21, 26] and postoperative pain after palmar plating has been underestimated. This randomized study demonstrated that patients who received a multimodal analgesic injection after palmar plating for a distal radius fracture had no clinically important reduction in pain scores, but they did consume lower doses of opioid analgesics, and they received anti-emetic drugs within 2 days of surgery.

This study has several limitations. The main limitation of the study is that our patients stayed in the hospital for 2 days and consumed considerably more opioids than patients in other countries who underwent outpatient palmar plating surgery. Our findings, however, imply that the amount of additional postoperative analgesia can be reduced by a local multimodal drug injection. Second, postoperative pain levels determined with a VAS score are subjective and might be influenced by psychological factors and personal experience. To overcome this limitation, we additionally used a fentanyl dose, which indicated the total pain intensity. Third, because we focused on the early postoperative period, we lacked data on follow-up pain levels or long-term clinical outcomes beyond 48 hours after surgery, which would have provided a better understanding of the temporal course of pain and its effect on functional outcomes. Fourth, pharmacokinetics and the effective dose of locally administered drugs for surgical-site multimodal injection have not been properly established, especially in surgery for DRFs. Because of this, we derived the approach we used from previous studies [9, 13-15, 20]. Thus, further study is necessary to find the optimal drug regimen for a multimodal drug injection. Fifth, there was the variation of the nursing staff who collected the VAS scores and administered fentanyl and antiemetic drugs. However, the operating surgeons, patients, ward nursing staff, and other data collectors remained blinded to the group status of patients for the duration of study. Finally, the current investigation did not assess the impact of this analgesic technique on elapsed operative time or cost, which may impact surgeons’ decisions to adopt this method.

A recent study reported that patients who received a brachial plexus block had increased pain, called rebound pain, between 12 and 24 hours after surgery [11]. Therefore, in the present study, all patients received palmar plating with general anesthesia to exclude the effect of a brachial plexus block and clearly identify the effect of the surgical-site drug injection. Although the multimodal injection group demonstrated lower VAS scores than the control group at 4, 8, and 24 hours postoperatively, a surgical-site multimodal injection did not result in a clinically important reduction in VAS pain levels after surgical fixation of DRFs. Chung et al. [9] performed a randomized controlled study of the use of a perifracture site injection in patients with palmar plating for DRFs. Similar to our results, peri-fracture site injections did not provide any additional pain control benefit. Most multimodal drug injection studies that found clinically important reductions in pain were usually performed for hip or knee arthroplasty. The mean postoperative VAS scores at 24 hours in the control group after these surgeries have been reported to range between 58 and 63 [14, 20], which were considered higher than the pain scores in the palmar plating for DRFs [9]. The postoperative VAS score encountered during the present study were also lower than the VAS scores reported in previous arthroplasty surgeries, because palmar plating is simpler and dissection or surgical times are shorter than those of total joint replacement surgery. For this reason, we believe that the effectiveness of a surgical-site multimodal drug injection for palmar plating might be underestimated.

The present study demonstrated that a ropivacaine-based anesthetic cocktail resulted in reduced early opioid consumption after palmar plating of DRFs. Opioids are commonly used to manage pain after surgery and are an important aspect of multimodal pain management, but opioids have well-known adverse effects that often make this option impractical and even dangerous [3]. One study reported that after open reduction and internal fixation of DRFs, 19% of orthopaedic surgery residents would prescribe more than 7 days of prescription opioids, which is beyond Massachusetts’s state law maximum [4]. This is especially important because the initial exposure to prescription opioids is thought to be a major risk factor for persistent use [10]. Reducing parenteral narcotic consumption is an overarching goals of multimodal pain management. Kang et al. [14] performed a randomized study of patients who received pre-emptive analgesic medication and an intraoperative surgical-site injection during hemiarthroplasty for hip fractures. They concluded that multimodal pain management provides additional pain relief until the fourth postoperative day, improves patient satisfaction at discharge, and reduces the total narcotic consumption for postoperative pain management. A recent randomized controlled study in patients with orthopaedic trauma also reported that a surgical-site multimodal drug injection improved pain control and reduced narcotic use during the first postoperative day, with no observed adverse effects [15].

In conclusion, our data suggest that patients who received surgical-site multimodal analgesic injections after palmar plating for distal radius fractures had no clinically important reduction in pain scores, but they did consume lower doses of opioid analgesics and fewer patients received anti-emetic drugs within 2 days of surgery. Opioids or other analgesia with high potency usually used for postoperative pain management have many side effects. Thus, reducing additional analgesia is as important as postoperative pain management and a surgical-site multimodal analgesic injection is one of the methods to achieve this a goal.