Comparison of the Efficacy and Safety of Dinalbuphine Sebacate, Patient-Controlled Analgesia, and Conventional Analgesia After Laparotomy for Gynecologic Cancers: A Retrospective Study

Introduction

Gynecologic cancers, including ovarian, endometrial/uterine, cervical, vulvar, and vaginal cancer, have an increasing trend in the female population aged over 50 years.¹ More than 80% of patients who underwent major operations experienced moderate and extreme postoperative pain (POP).^2,3 Poorly controlled POP causes physiological and psychological consequences and leads to risk of morbidity, complications, delayed wound healing, prolonged hospitalization, and psychological disorder from the traumatic stress.⁴

Enhanced recovery after surgery (ERAS) has been developed with the aim of maintaining normal physiology during the perioperative period, leading to improved clinical outcomes. Multiple studies have demonstrated that the implementation of ERAS was associated with reduced length of hospitalization, lower postoperative complication and readmission rates, and accelerated postoperative recovery in various surgical areas, including gynecologic, colorectal, gastrointestinal, orthopedic, vascular, and hepatobiliary area.^5–9 Appropriate pain management played a significant role in achieving ERAS. Currently, a multimodal opioid-sparing regimen involved in multiple drug combinations through different mechanisms of pain control that act synergistically while limiting opioid consumption is the mainstay for perioperative pain management.¹⁰

Several approaches have proven to show effective control of POP, including conventional analgesia (CA) given as needed, patient-controlled analgesia (PCA), or pre-emptive analgesia such as extended-release dinalbuphine sebacate (DNS).^11–13 Although CA could decrease the pain intensity in several minutes, the incidence of adverse events (AEs) was usually significant and even intolerable to patients in the early recovery period.¹⁴ Patients experience POP before getting medication and psychological stress from making frequent calls for medical care. Systemic intravenous (IV) PCA with opioids such morphine or fentanyl plays an important role in POP control; however, its debilitating side effects such as sedation, nausea, dizziness, and pruritus may lead to delayed postoperative rehabilitation.¹⁵ DNS (Naldebain^® ER Injection, Lumosa Therapeutics Co. Ltd., Taiwan) was developed as a long-acting analgesic that maintains the nalbuphine plasma level at the therapeutic range of up to 7 days through intramuscular (IM) administration. It has shown benefits of less postoperative analgesic consumption and sufficient efficacy in POP pain intensity management.^12,16,17

In this study, we retrospectively analyzed the clinical data of patients with gynecologic cancer who received laparotomy through a midline incision in a single center. The primary objective of this study was to compare the efficacy, safety, and postoperative indicators between CA, PCA, and DNS for critical recommendation in POP management integrated with ERAS in the future.

Methods and Designs

Setting and Participants

The clinical data of 147 patients with gynecologic cancers who underwent laparotomy in our hospital between July 1, 2019, and June 1, 2020, were retrospectively collected. The study protocol was reviewed and approved by the institutional review board of Chang Gung Memorial Hospital (IRB202000594B0) and conducted in accordance with the Declaration of Helsinki. Source data were collected from the Hospital Information System in the department of obstetrics and gynecology of Chang Gung Memorial Hospital. We only collected the clinical data of patients retrospectively and did not interfere with treatment; hence, individual consent for this retrospective analysis was waived. The 137 eligible patients aged between 20 and 80 years who underwent laparotomy through a midline incision for gynecologic cancer were evaluated in this study. The patients received general anesthesia, with American Society of Anesthesiologists (ASA) physical status classes 1 to 3. Ten patients were excluded for any of the following situations: incomplete POP evaluation with a numerical rating scale (NRS), second-look laparotomy, receiving nerve block for pain control, allergy to local anesthetic agents or morphine, beyond the recommended DNS administration time period (12 to 24 hours before surgery), severe comorbidity (cardiopulmonary disease and cerebrovascular accident), or chronic use of morphine.

Analgesic Regimen

The 137 eligible patients in this study were categorized into three groups according to the method of postoperative analgesia as follows: DNS, intravenous PCA with fentanyl, or CA. Fifty-two patients in the DNS group received a single dose of IM DNS (150 mg/² mL) using the Z-track technique into the buttock area 12 to 24 hours before surgery. In Z-track injection, the epidermal and hypodermal layers of the subcutaneous tissue were moved 2 to 3 cm to the side by the nondominant hand while performing the deep muscle injection in the gluteus maximus in the buttocks by the dominant hand.¹⁸ Fifty-five patients in the PCA group received IV PCA with fentanyl from recovery through 72 hours after surgery. The PCA device was programmed to administer a bolus dose of 16.56 μg fentanyl (3.125 μg/mL, 5.3 mL) at an infusion rate of 4.2 mL/hr. The mean lockout interval was 5.3 min. The basal infusion rate was 1.5 mL/hr in the PCA group. If PCA with fentanyl was deemed unnecessary by patients, the PCA machine could be removed 48 hours after surgery. In CA regimens, 30 patients received IV parecoxib 40 mg within 3 hours before the surgery, in accordance with the internal guideline in Chang Gung Memorial Hospital. Patients with NRS scores of 5 or above from any group may receive rescue analgesic medications, including IM nalbuphine 10 mg, IV parecoxib 40 mg, IV fentanyl 25/50 mcg, IV morphine 2–10 mg, and IV ketorolac 30 mg. Fentanyl, morphine, and ketorolac were administered as single-agents in the post-anesthesia care unit at the discretion of anesthesiologist. In the ward, patients received single agent nalbuphine or parecoxib each time when needed according to their personal choice. The quantity and frequency of the rescue medications were recorded in this study.

Pain Assessments

All the eligible subjects were evaluated for POP using the NRS grading from 0 to 10, where 0 represents “no pain” and 10 represents “unbearable pain.” Serial monitoring of the NRS scores was performed three times a day at 8-hour intervals (8:00 am, 2:00 pm, and 8:00 pm) for six consecutive days (from the operation day to postoperative day 5 or the discharge day). The daily representative NRS score in each patient was analyzed by averaging three NRS scores each day in each patient. The mean NRS score indicated the average of the daily representative NRS score of the patients in the DNS, PCA, or CA groups. Day 0 represents the operation day. Days 1 to 5 represent the first day to the fifth day after operation, respectively.

Endpoints

The endpoints were comparisons of the differences in NRS pain scores from day 0 through day 5, total postoperative morphine equivalents per day, time of ambulation, and first time of flatulence between the different groups. Treatment-emergent AEs (TEAEs) were observed and recorded from the time of investigational drug administration through day 5 or discharge day. The severity of each AE was assessed using the National Cancer Institute-Common Terminology Criteria for Adverse Events (NCI-CTCAE)

Data Analysis

Statistical analysis was performed using the SPSS software package. Results were presented as mean ± SD. Student’s t-test was used for numerical variables, and chi-square test or Fisher exact test was used for categorical variables. The consumption of morphine equivalent dose was calculated by applying the conversion ratio of fentanyl to morphine as 0.01 to 1 and nalbuphine to morphine as 1 to 1.¹⁹ One-way analysis of variance (ANOVA) was performed to compare the difference in the consumption of analgesics among the DNS, PCA, and CA groups. Post hoc analysis using Dunnett’s multiple comparisons test was conducted to examine differences by comparing DNS to CA and DNS to PCA. Statistical significance was defined as p < 0.05.

Results

Patient Clinical Characteristics

During the study period, a total of 147 patients underwent laparotomies for gynecologic cancers, and 137 patients were included in the study after excluding those who did not meet the inclusion criteria. Fifty-two patients were included in the DNS group; 55 in the PCA group; and 30 in the CA group (Figure 1). Demographics and baseline characteristics were summarized in Table 1. The patients’ characteristics, including age, weight, body mass index, incision site, postoperative length of hospital stay, and ASA score, were similar between the groups. The patients were diagnosed as having ovarian, uterine, or cervical cancer. Among the 137 patients, the percentage of patients with ovarian cancer was relatively higher in the DNS group than in the PCA or CA group (Table 1).

Table 1 Patient Characteristics

Figure 1 Flow chart of participants in this study.

Efficacy Outcomes

POP was measured from day 0 to 5 using the NRS. The DNS group showed significantly lower NRS scores than the PCA and CA groups on day 1 (4.8 vs 6.2, p < 0.01 and 6.2, p < 0.05, respectively), day 2 (3.0 vs 4.7, p < 0.01 and 4.8, p < 0.001, respectively), day 3 (2.0 vs 3.9, p < 0.001 and 3.5, p < 0.001, respectively), day 4 (1.1 vs 3.1, p < 0.001 and 2.9, p < 0.001, respectively), and day 5 (0.7 vs 2.3, p < 0.001 and 2.4, p < 0.001, respectively; Figure 2). Meanwhile, the DNS group showed a significantly shorter time to ambulation than the PCA group (2.5 ± 1.3 days vs 3.2 ± 1.4 days; p = 0.008), similar time to ambulation when compared with the CA group (2.5 ± 1.3 days vs 2.0 ± 0.9 days, p = 0.052). In addition, DNS group showed no significant difference in time to flatulence when compared with the PCA group (3.0 ± 1.4 days vs 3.0 ± 1.1 days, p = 0.979), but showed longer time when compared with the CA group (3.0 ± 1.4 days vs 2.1 ± 1.0 days, p = 0.005). The total consumption of morphine equivalents per day was similar between the DNS and PCA groups (142.8 ± 7.3 mg vs 137.7 ± 70.0 mg, p = 0.8032) and lowest in the CA group (11.7 ± 30.7 mg, p < 0.0001; Table 2). The parecoxib dose was significantly higher in the CA group than in the DNS group (118 ± 121.6 mg vs 30.8 ± 64.2 mg, p < 0.0001) but was not significantly different between the DNS and PCA groups (30.8 ± 64.2 mg vs 33.5 ± 36.7 mg, p = 0.9745). In addition, no significant difference in ketorolac dose was found among the three groups.

Table 2 Consumption of Analgesics

Figure 2 Post-operative pain assessment in patients receiving DNS, PCA or CA from day 0 to day 5. Notes: *p<0.05, **p<0.01, ***p<0.001.

No significant difference was found in fentanyl dose between the DNS, PCA, and CA groups either during the induction (79.8 ± 29.7 mg vs 72.7 ± 27.0 mg vs 69.2 ± 24.3 mg, p = 0.198) or maintenance phase (115.2 ± 49.3 mg vs 121.1 ± 53.5 mg vs 120.8 ± 61.5 mg, p = 0.830; Table 3).

Table 3 Intraoperative Consumption of Fentanyl

Safety and Adverse Effects

The incidence rates of TEAEs are presented in Table 4, and all TEAEs were CTCAE grade 1/2 in severity. In the DNS group, 14 patients (27%) experienced dizziness; 10 (19%), nausea; 9 (17%), vomiting; and 6 (12%), pain at the injection site. Other than pain, no local erythema or infection over the injection site was observed. In the PCA group, the most common TEAE was dizziness, found in 26 patients (47%), followed by nausea in 19 patients (35%) and vomiting in 15 patients (27%). In the CA group, 10 patients (33%) complained of nausea and 8 patients (27%) complained of vomiting. Overall, the safety profile was comparable between the DNS, PCA, and CA groups. A trend of higher dizziness in the PCA group than in the DNS and CA groups was observed. The incidence rates of nausea, vomiting, and retching were lowest in the DNS group.

Table 4 Summary of Incidence of Treatment Emergent Adverse Events (TEAEs) Reported in Each Study Group

Discussion

This retrospective study demonstrated that preemptive administration of DNS in patients with gynecologic cancer who underwent laparotomy is an effective and safe approach for POP management. To our knowledge, this is the first study to evaluate the efficacy of pain relieving and safety of DNS in patients with gynecologic cancer who underwent laparotomy and to compare PCA with fentanyl and CA regimens in real-world practice. Significantly lower NRS pain scores from day 1 to 5 were observed in the patients who received a single dose of 150-mg IM DNS injection 12 to 24 hours before surgery as compared with those in the PCA or CA group. Consumption of postoperative morphine equivalent was similar between DNS and CA group, but was lower in the DNS group when compared with the PCA group. The overall safety profile was comparable between the DNS, PCA, and CA groups. In addition, DNS showed a significantly shorter time to ambulation than the PCA group.

Gynecologic operations, especially laparotomy with a midline incision for gynecologic cancer, cause severe pain. The commonly used intravenous opioids are IV opioids, morphine and fentanyl, owing to their rapid onset of action with peak effect within 1–2 hours after administration.²⁰ Fentanyl is a synthetic opioid that is 50–100 times more potent than morphine.²¹ Continuous IV PCA was first introduced in the 1970s. By using a computer-controlled pump, patients are able to obtain additional dose of the drug according to the pain on the spot by themselves.¹⁵ Being a long-acting nalbuphine, a single-dose, preemptive IM injection of DNS has been reported to result in a significant reduction of post-hemorrhoidectomy pain to as long as 7 days after operation.¹⁶ Furthermore, a randomized controlled study reported that a single dose of DNS was superior to intravenous PCA with fentanyl in patients who underwent laparotomy.¹² In the present study, the patients received laparotomic staging, debulking, or radical surgery for uterine, ovarian, or cervical cancers. For patients with suspected ovarian cancer, intraoperative frozen section diagnosis was performed during surgery. In addition, some of the selected patients underwent further hyperthermic intraperitoneal chemotherapy (HIPEC) during the surgery. As the above-mentioned procedures required additional operation time, the average operation duration was relatively long (5.9–6.8 hours). Besides, a higher proportion of patients were diagnosed as having ovarian cancer in the DNS group, which resulted in a higher proportion of patients receiving debulking or HIPEC and longer operation time.²² These complicated operative procedures might lead to severe POP and longer recovery time in the DNS group. In our study, a trend of lower POP intensity in the DNS group than in the PCA or CA group was observed, which might further support the evidence of the potential analgesic efficacy of DNS for POP management in patients with gynecologic cancer.

Appropriate perioperative patient care plays a critical role in ERAS.^23,24 Numerous studies reported that multimodal and preemptive analgesia for POP management attained early mobilization, reduced discomfort, and promoted faster recovery.^25–27 Our study showed that the patients who received preemptive DNS exhibited significantly faster ambulation than those who received PCA group, indicating the beneficial effect of DNS for promoting early recovery in postoperative patient care. Meanwhile, no statistical significance was found in the days to ambulation between the DNS and CA groups. As IV PCA and CA both require a pain management team to conduct POP control, preemptive IM injection of DNS may have a potential to alleviate the burden of the site staff while providing sufficient pain relief.

Owing to its short half-life, nalbuphine requires frequent injection in clinical practice to achieve adequate POP control, resulting in greater incidence of AEs such as sedation, dizziness, nausea, and vomiting.²⁸ Tien et al previously demonstrated DNS as an extended-release nalbuphine that takes approximately 6 days to completely release into the bloodstream.²⁹ When released, DNS can be rapidly hydrolyzed to nalbuphine by esterase to exert its analgesic effect. Therefore, with a lower maximum blood concentration, DNS has a better safety profile than nalbuphine itself. No opioid antagonism or severe nalbuphine-induced AEs were observed in the DNS group, which may due to the single-dose extended release of DNS resulting in a relatively low maximum plasma concentration of nalbuphine (C_max= 16.8 ng/mL),³⁰ which led to the lower trend in dizziness, nausea, and vomiting observed in our study. While injection site reactions (ISRs) such erythema and swelling were commonly reported AEs in oil-based IM injection products.³¹ In this study, 12% of the patients experienced pain, with no local erythema or infection over the injection site. Compared with a previous study that used IM DNS injection for POP control, the present study had a lower incidence of ISR, that is, 27.5% with erythema or swelling at the injection site versus 12% with pain.¹⁶ We believe that this difference was due to the use of the Z-track technique. Applying the Z-track technique for IM injection has shown to reduce drug leakage to the subcutaneous tissue.¹⁸ Our study also supports the use of the Z-track technique to lower the ISR incidence rate in patients receiving DNS.

As DNS is a kappa opioid receptor agonist and mu opioid receptor antagonist, the strong binding affinity between the mu receptor and DNS may lead to increased fentanyl dose for anesthesia induction. In our study, we found no significant difference in fentanyl dose between the DNS, PCA, and CA groups either during the induction or maintenance phase.

Regional anesthesia such as peripheral nerve blocks, neuraxial anesthesia, and local anesthesia infiltration reported in recent studies showed advantages in safety, such as early ambulation and recovery, shorter length of hospital stay, reduced postoperative nausea and vomiting, and improved pulmonary outcomes.^32–34 Further study should be conducted to evaluate and compare the efficacy and safety between DNS, regional, and conventional anesthesia and examine the potential of integrating DNS to regional combined with conventional analgesic procedures.

The study has some limitations that should be mentioned. First, our study was performed in a single center with 30–55 patients in each arm. In addition because the study was retrospective in design, patient demographics was not well balanced and the proportion of patients with ovarian cancer was higher in the DNS group. Second, the pain score was self-rated by the study participants, and the differences in pain threshold and pain tolerance level varied among individuals, which might have caused the discrepancy. The observed effective analgesic ability of DNS may provide evidence of the feasibility of incorporating DNS as part of perioperative pain management in the future.

Conclusion

Our real-world data showed that a single dose of DNS relieved POP effectively in the patients who underwent a longitudinal abdominal incision for gynecologic cancers and exerted superior pain control as compared with PCA or CA. Further prospective randomized studies are still needed to obtain more information on and firmer evidence of the effects of the drug.