The review search methodology identified studies reflecting the role of opioids in pediatric oncological surgery in a variety of surgical settings: thoracic surgery [8, 9, 52], abdominal surgery [10,11,12,13,14,15,16, 53], orthopedic surgery [17,18,19], neurosurgery [20, 21], and ophthalmology surgery [22]. Several studies focused on the role of epidural analgesia and the use of opioids perioperatively, for various types of oncological surgery [23,24,25,26], while others evaluated interventions for opioid reduction perioperatively [27,28,29].

3.1 The Role of Opioids for Thoracic Surgery in Pediatric Oncology

Three single-center retrospective studies reviewed post-thoracotomy pain control in children with Ewing sarcoma, osteosarcoma, and other cancer diagnoses [8, 9, 52]. Of these studies, two [8, 52] investigated the use of intercostal nerve cryoablation (INC) to reduce pain and opioid use, and one [9] compared the use of patient-controlled analgesia (PCA) versus epidural analgesia for pain management (Table 1).

Table 1 Opioids in pediatric oncology patients undergoing thoracic surgery

Two studies reviewed the analgesic benefit of INC when compared with standard treatment, with the primary outcome of total opioid usage postoperatively. Specific opioid dosages were not provided in these studies; cumulative opioid consumption was reported as total morphine milligram equivalents (MME) per kilogram [8] and total oral morphine equivalents (OME) [52]. Both studies found that the overall opioid consumption was lower in the groups treated with INC for post-operative pain management. Postoperative opioid usage after thoracotomy and the need for opioids at discharge were also evaluated in these studies and found that patients who received INC were less likely to receive opioids upon discharge than those without INC [8, 52].

In addition to opioid consumption, authors reported pain scores postoperatively, as an analgesia quality outcome, in comparisons of groups treated with INC versus standard treatment [8] and in epidural versus PCA analgesia [9]. When comparing total pain scores measured, Chen and colleagues found there were no significant differences (p = 0.20) between groups; however, when evaluated by day, pain scores became significantly different on postoperative day (POD) 4 with a median maximum pain score in the INC group of 2.5 versus 5 for the non-INC group (p = 0.01) [8]. Gonzalez and colleagues found the opposite—pain scores during the overall postoperative period were significantly different while pain scores on POD 1 and POD 2 were similar between groups [9].

3.2 The Role of Opioids for Abdominal Surgery in Pediatric Oncology

Seven studies were identified that had data regarding the outcome of opioid utilization for pain management after abdominal surgery for pediatric oncology: three studies included diverse types of intraabdominal cancer surgeries [10,11,12], one pertained to Wilms tumor surgery [13], and three addressed the unique surgical challenge of cytoreductive surgery (CR) with hyperthermic intraperitoneal chemotherapy (HIPEC) [14,15,16]. The subsequent re-run of the literature search completed in March 2024 retrieved an additional reference, relevant to neuroblastoma surgery [53] (Table 2).

Table 2 Opioids in pediatric oncology patients undergoing abdominal surgery

One of the earliest retrospective studies of pain management after abdominal surgery investigated the value of epidural analgesia for abdominal surgery, by comparing outcomes in 40 patients in the study group (mean age 4.65 years) and 44 in the control group (mean age 3.25 years), with diagnoses of ganglioneuroma, rhabdomyosarcoma (RMS), neuroblastoma, pheochromocytoma, nephroblastoma, focal nodular hyperplasia, hepatoblastoma, hepatocellular carcinoma, and germ cell tumor, based on pair-matched historical controls. Children treated with epidural analgesia received significantly fewer doses of piritramide or morphine (45% vs 82%, p < 0.001) and had lower pain intensity scores on POD 1 and 3 compared with the control group, though the difference was not statistically significant (p = 0.15 and 0.09, respectively). Differences in clinical outcomes including mechanical ventilation time, time in intensive care unit, and total hospital stay were not statistically significant between groups [10].

Two randomized controlled trials (RCTs) investigated distinct intraoperative analgesia protocols while utilizing either postoperative opioid consumption [11], or pain scores and time to first rescue analgesia as outcome measures [12]. In a randomized controlled study of three equal groups, 90 children (7–12 years old, demographically comparable between groups) undergoing major abdominal surgery for cancer through a midline incision were treated either with a single dose of intravenous (IV) morphine 0.1 mg/kg after induction of anesthesia or caudal epidural block (bupivacaine and morphine) or a multimodal protocol of acetaminophen, ketorolac, IV ketamine infusion, and local anesthetic wound infiltration before skin incision and at the end of surgery. All groups used morphine PCA for postoperative pain management, at doses of 20 μg/kg with lockout of 10 min. The multimodal and caudal groups did not significantly differ in terms of total morphine consumption (p = 0.521), time to first analgesic bolus (p = 0.136), or the number of total and active PCA boluses (p = 0.260 and p = 0.904, respectively). Both groups were superior to the morphine group regarding morphine consumption, time to first analgesic bolus, and the number of total and active PCA boluses. Total morphine consumption doses via PCA (mg/24 h), as mean and standard deviation (SD), were morphine group, 12 ± 2, and caudal and multimodal group, 6 ± 1 (p < 0.001) [11].

Another RCT of 60 pediatric oncology patients, aged 3–12 years, undergoing major abdominal surgery utilized three groups (20 patients each) treated with intrathecal medications: bupivacaine alone (control group), bupivacaine plus fentanyl or bupivacaine plus dexmedetomidine. The fentanyl and dexmedetomidine groups demonstrated significantly lower mean FLACC scores (Face, Legs, Activity, Cry, Consolability) at 6, 8, and 12 h postoperatively versus control (p = 0.05). Time to first analgesic request was significantly delayed, and acetaminophen utilization was significantly decreased in the dexmedetomidine group, suggesting that overall, the addition of dexmedetomidine versus fentanyl to standard intrathecal bupivacaine may provide analgesic advantages. This study did not report opioid consumption postoperatively as an outcome measure; rather, it utilized the addition of opioids to an intrathecal route of analgesia in comparison with standard local anesthetic or local anesthetic and dexmedetomidine [12].

A specific focus on unilateral nephrectomy for Wilms tumor with (n = 46) or without epidural analgesia is reflected in a retrospective study of 77 children, reporting postoperative opioid consumption as OME/kg, receipt of opioid prescription at discharge, and length of stay (LOS). Patients treated with epidural analgesia used significantly lower amounts of opioid in hospital, though there was no significant difference in opioid discharge prescriptions or overall postoperative LOS (median days 5 vs 6; p = 0.10) [13].

Following the trend of evaluating the outcomes of standardized perioperative enhanced recovery practices (ERPs), Lee et al. retrospectively evaluated 37 pediatric patients, median age 33 months (IQR: 20–48 months), undergoing open resection of abdominal high-risk neuroblastoma (HR-NB), regarding the outcomes of two perioperative ERPs: avoidance of routine nasogastric tube (NGT) use, and use of neuraxial anesthesia. In this context, the primary outcomes of time to enteral intake, urinary catheter use, opioid utilization, and LOS were reported. The utilization of neuraxial anesthesia in 29 patients (78%) (27 with epidural catheters and 2 with spinal or caudal blocks) was associated with lower OME/kg on POD 0 (0 mg/kg vs 0.17 mg/kg; p = 0.05), yet there was no difference in OME consumption on subsequent days or cumulatively for the postoperative hospital course (p = 0.77). Neuraxial anesthesia was frequently associated with indwelling urinary catheter (p < 0.01) for a longer duration (p = 0.002), and postoperative LOS was not influenced by use of neuraxial anesthesia (p = 0.89), leading to the conclusion that neuraxial anesthesia necessitates the placement of a urinary catheter without decreasing opioid utilization, making its benefits less evident (that is, avoiding the use of an NGT which led to earlier feeding, without influencing the LOS) [53].

Three studies generated data regarding opioid utilization in the context of the CR HIPEC procedures, recognizing that pain management for this complex type of surgery is challenging [14,15,16]. In a retrospective investigation of the impact of opioid consumption on the survival of 75 children and adolescents aged < 19 years (43% female) undergoing CR with HIPEC, for diagnoses of desmoplastic small round cell tumor (DSRCT) (n = 35), RMS (n = 13), colorectal (n = 4), mesothelioma (n = 3), or others (n = 20), the authors described a median perioperative (cumulative intraoperative and postoperative) opioid consumption, expressed as weight-adjusted IV morphine equivalent dose (MED), of 18.9 MED/kg (range 0.6–339.6). Intraoperative analgesia was supported with continuous epidural analgesia and continuous infusion of fentanyl or sufentanil, followed by PCA with morphine, hydromorphone, or fentanyl postoperatively. Their analysis found no statistically significant association between opioid consumption and recurrence-free survival (hazard ratio [HR] 1.00; 95% confidence interval [CI] 0.99–1.02; p = 0.55) or overall survival (HR 1.01; 95% CI 0.99–1.03; p = 0.22). Independent prognostic factors associated with poor survival included incomplete CR and extra-abdominal disease [14].

A retrospective analysis reported on nine patients undergoing CR HIPEC surgeries for DSRCT, median age 19 years, seven males, for which epidural infusions were utilized in eight cases for a median of 4 days (range 3–5), followed by postoperative pain management with PCA with morphine or hydromorphone. The postoperative IV opioid use (morphine equivalent) was 0.67 mg/kg/day (range 0.1–9.2), administered for a median of 11 days (range 2–35) [15]. Another retrospective analysis of 25 patients undergoing CR HIPEC, most commonly for DSRCT (n = 12), median age 14 years, used a protocol consisting of premedication with gabapentin, intraoperative infusions of propofol, dexmedetomidine, and ketamine, and thoracic epidural infusion analgesia for a median duration of 8 days. Median postoperative IV opioid use (morphine equivalent) through POD 10, including opioid PCA initially and oral opioids subsequently, was 0.02 mg/kg/day, with median duration of 2 days. Based on the low reported median opioid consumption and the fact that nine patients (36%) did not require any IV opioids in the postoperative period, the authors concluded that this multimodal intraoperative/postoperative analgesia can reduce opioid exposure [16].

3.3 Long-term Opioid Reduction Studies in Oncological Surgery

The search strategy identified three studies with a specific focus on interventions to reduce long-term opioid exposure after pediatric oncological surgery, one evaluating the risk of persistent opioid use in the context of abdominal surgery [27], and the others describing the quality improvement (QI) process for development [28] and subsequent validation of an opioid prescribing predictive algorithm after abdominal and thoracic oncological surgeries [29].

A retrospective investigation of the prevalence and factors associated with the development of persistent opioid use up to 6 months postoperatively, in 86 children and adolescents, median age 12 years (range 2–19), after CR HIPEC for cancer diagnoses of colorectal 4 (5%), DSRCT 40 (47%), mesothelioma 3 (3%), RMS 14 (16%), and others 25 (29%), analyzed the opioid exposure immediately postoperatively as morphine (0–0.23 mg/kg/h), fentanyl (0–0.05 μg/kg/h), or hydromorphone (0–0.004 mg/kg/h), by IV PCA, with demand doses of morphine (up to 0.02 mg/kg), fentanyl (up to 0.5 μg/kg), or hydromorphone (up to 0.004 mg/kg), and additional nurse-administered boluses of morphine (0.5–4 mg/h) as needed. The proportion of patients with subsequent persistent opioid use at 3, 6, 12, and 24 months post-discharge was 70%, 35%, 29%, and 13%, respectively. Both higher average daily inpatient pain intensity scores (estimated difference 0.5, 95% CI 0.3–0.8; p < 0.01) and higher postoperative opioid consumption (odds ratio 1.03, 95% CI 1.00–1.07; p = 0.05) were associated with persistent opioid use of up to 6 months. Specifically, the postoperative opioid consumption was described as mean (SD) MED/kg 26.4 (43.0) and median (IQR) 12.8 (5.5–29.3) for the overall group, and as mean (SD) for the groups without persistent opioid use versus persistent opioid use as 14.5 (13.5) versus 25.8 (24.0); p = 0.05 [27].

A QI initiative to minimize opioid prescribing after oncologic pediatric surgery consisted of targeted education regarding nonopioid analgesia and tailored opioid prescription at discharge based on retrospective prior inpatient opioid requirements (n = 271), with prospective data collection on 99 patients. Mean (SD) OME/kg prescribed upon discharge was significantly reduced in the prospective (0.75 ± 1.34) versus retrospective cohorts (5.48 ± 6.94; p < 0.001). The unplanned visits/calls regarding pain were 23 (retrospective, 8.5%) and 2 (prospective, 2.0%). In total, 44 patients (44.4%) received an opioid prescription at discharge in the prospective cohort, significantly fewer than the retrospective cohort (251, 92.6%, p < 0.001), and used a mean of 34.3 of 159.8 (21.5%) doses dispensed. Length of stay was comparable (p = 0.88) between cohorts. The prospective satisfaction rate was 96.2%, leading the authors to conclude that a dramatic reduction of opioid prescriptions after oncologic surgery can be achieved without detriment to patient satisfaction or readmissions [28].

A prospective QI study further pursued validation of an opioid prescribing predictive algorithm after abdominal and thoracic oncological pediatric surgeries. The prescribing algorithm was developed based on the surgical approach, POD of discharge, and inpatient opioid use. Data prospectively collected included outpatient opioid consumption and patient/family satisfaction. The total home dose prescribed at discharge was equal to that used in 8 or 24 h, depending on the LOS and operative approach before discharge, divided into 0.15-mg/kg doses. Of 121 patients, the algorithm correctly predicted outpatient opioid requirements for 102 patients (84.3%), over-estimated opioid need in 15 (12.4%) patients by an average of 0.38 OME/kg and under-estimated the need in four (3.3%) patients who required additional opioids. The utilization of the algorithm led to the reduction of overall opioid prescriptions from 6.17 to 0.21 OME/kg (p < 0.001), pre- and post-QI intervention, and a single patient/family reported dissatisfaction with postoperative pain control, leading to the conclusion that outpatient opioid needs can be accurately predicted, ensuring avoiding excess opioid prescriptions, without compromising patient satisfaction [29].

3.4 The Role of Opioids in Pediatric Oncology Patients Undergoing Orthopedic Surgery

Three retrospective studies were found that describe the role of opioid analgesics for acute perioperative pain in pediatric cancer patients undergoing orthopedic surgery (Table 3) [17,18,19]. In one study, a multimodal pharmacologic analgesia regimen that included fentanyl, hydromorphone, or methadone allowed for acceptable pain control after hemipelvectomy in pediatric osteosarcoma and Ewing’s sarcoma patients (n = 8). Mean oral MED/kg peaked on POD 5 (5.03 ± 6.06) while mean pain scores were highest on POD 0, POD 5, and POD 30 [17]. The authors noted that pain on POD 0 is likely secondary to the surgery and that the peak opioid administration on POD 5 coincided with the observed mean time to ambulation of 5.33 ± 2.94 days.

Table 3 Opioids in pediatric oncology patients undergoing orthopedic surgery

A second retrospective study evaluated patterns of outpatient opioid prescribing patterns in 28 pediatric patients with bone sarcoma after primary tumor resection [18]. Patients with preoperative opioid use were prescribed significantly higher opioid doses at all time points post-discharge compared with opioid-naive patients. Pre-operative opioid use was directly associated with higher 120-day total opioid requirements. Patients with pelvic tumors had greater postoperative opioid use than those with upper or lower extremity tumors during the 61–90-day and 91–120-day intervals post-discharge. Presence of metastasis was associated with higher opioid prescriptions in the first 30 days after surgery. Increasing age, diagnosis of osteosarcoma, longer hospital stay, and presence of metastasis were factors linked to increased 30-day postoperative opioid requirements [18].

Argun and colleagues evaluated the efficacy of ultrasound-guided peripheral nerve blocks for postoperative analgesia in 108 pediatric orthopedic tumor surgery patients. Patients who received peripheral nerve blocks with bupivacaine 0.25% had lower total analgesic (tramadol) consumption in the first 24 h postoperatively as compared with the group who received only IV analgesics. Similarly, pain scores were significantly lower in the peripheral nerve block group at 0, 1, 2, and 6 h postoperatively, and the mean time to pain onset was significantly longer in the peripheral nerve block group [19].

3.5 Opioids and Epidural Analgesia in Pediatric Oncology Patients

Four studies focused on managing postoperative pain in pediatric oncology patients receiving epidural analgesia [23,24,25,26] (Table 4). These studies employed a range of opioid agents and dosing regimens. Overall, the use of post-operative epidural analgesia in pediatric oncology patients appears to be effective in controlling pain and led to lower overall opioid consumption.

Table 4 Opioids and epidural analgesia in pediatric oncology patients

One study retrospectively reviewed the experience with epidural analgesia in 58 pediatric oncology patients over 2 years [23]. Fentanyl was administered as an epidural infusion at 1 μg/kg/h, following a bolus dose of 1 μg/kg. The epidural catheter was removed on POD 3, and analgesia was subsequently provided as needed with oral acetaminophen or codeine/oxycodone. The authors reported that epidural analgesia was a safe and effective means of postoperative analgesia after major surgical procedures.

The safety of concurrent use of epidural and IV opioids in 117 pediatric oncology patients was examined retropectively [24]. Epidural opioids included fentanyl as a continuous infusion of 0.5–1 μg/kg/h and doses of IV opioids were administered as needed as fentanyl 0.5–1 μg/kg, morphine 0.05–0.1 mg/kg, or hydromorphone 0.01–0.02 mg/kg. Alternatively, opioid PCA was utilized, with morphine as the first-line agent. Most patients had mild pain scores (0