DISCUSSION
Similar to prior publications in pediatric hematology/oncology patients, we were able to show that patients initiated on LDKI tolerated the infusion without significant side effects resulting in LDKI discontinuation. Patients utilized LDKI for a median of 6 days (IQR 4, 11) also signifying tolerability. Also similar to prior publications, LDKI did not impact opioid utilization. Although we did not show that there was a significant impact on pain scores, there may have been a clinically meaningful, subjective reduction in pain. Overall, there was a 30% reduction in median pain scores when comparing the entire cohort on the day of LDKI initiation (day 0) and the subsequent day (day 1). Similarly, when comparing patients with sickle cell disease receiving LDKI to prior admissions without LDKI, there was a 32% reduction in median pain scores, though this was underpowered to reach significance.
Studies attempting to define clinically meaningful pain reduction are limited. Generally, a 30%-50% decrement in pain scores is considered clinically meaningful, though cannot be generalized to the individual patient. [24] In studies in both acute, post-operative pain as well as chronic pain, a decrement in the visual analog scale of 20% corresponded to minimal improvement, while 30%-35% decrement corresponded to much improvement. [25-26] Myrvik et al. [27] note in a small pediatric SCD cohort that a decrement in the visual analog scale of 0.97 cm and 0.9 in the numeric rating scale was the minimum clinically significant improvement. Numerical pain rating scales are limited due to being a unidimensional assessment of pain at one moment in time in patient cohorts with dissimilar pain tolerance and medication response. [28] As noted by Nobrega et al. [18] in their study of pediatric patients with sickle cell disease, there are significant inter-patient differences in response to LDKI; therefore, there may be a subset of patients more likely to benefit but not yet clearly delineated.
Prior studies with LDKI in pediatric hematology/oncology patients (Table 1) have similarly struggled to define an objective response for multiple reasons: 1) studies were almost all retrospective; 2) patients with end-stage cancer had increasing pain due to their underlying condition; 3) patients’ pain may have naturally improved over time and improvement unnecessarily attributed to LDKI initiation (i.e., regression to the mean); 4) multiple different measures were utilized to define which pre-LDKI and post-LDKI pain scores to compare; and 5) challenges of generalizing an intervention with significant inter-patient variability. The single recent multicenter prospective observational trial by Courade et al. [9] in pediatric and young adult cancer patients reported meaningful benefit in pain scores when comparing the day of LDKI initiation and two days later. Given the rapid onset of action of LDKI, two days may or may not be an appropriate time frame to measure response. More important though, 47% of patients found LDKI to be “very helpful,” which further argues that the objective response was real, at least for some of the patients. Nobrega et al. [18] reported a meaningful reduction in pain scores in a large cohort of pediatric patients with sickle cell disease, but this was comparing baseline pain scores with pain scores taken after LDKI discontinuation, which was several days later and could be attributed to the natural improvement in patient symptoms. The two retrospective studies of patients with chronic pain reported improvement in pain scores immediately following LDKI which is the more pathophysiological expected time frame for response given the rapid onset of action of LDKI. [22,23]
Additionally, the best dose of LDKI is yet to be defined. In our protocol we set a dose range of 1-10 mcg/kg/min (0.06-0.6 mg/kg/h) which is similar to the doses used in the majority of the other reported pediatric studies. Given we reviewed patients after protocol initiation, the majority of patients were started on the low end of the dose range given the median dose of LDKI was 0.06 mg/kg/h (1 mcg/kg/min). Whether patients may have had a more meaningful response at higher doses is not clearly delineated by our or other studies. The single patient that required dose reduction in our retrospective analysis was on 2 mcg/kg/min (0.12 mg/kg/h) and reduced to 1 mcg/kg/min (0.06 mg/kg/h). Coubarde et al. [9] similarly note that patients with side effects were all on lower doses (<0.05 mg/kg/h). Our study has multiple limitations including: 1) being a retrospective analysis; 2) having a small, heterogenous cohort and 3) lacking an a prioridefinition of best pain scores to compare and what entails a clinically meaningful result.
Since initiation of our LDKI protocol, empirically we have seen that certain patients with sickle cell disease and chronic pain perceive benefit with LDKI while others do not, not dissimilar to certain patients who prefer hydromorphone to morphine and vice versa. We have also empirically seen that many patients with chronic pain and at end of life are maintained on higher doses (i.e., 3-10 mcg/kg/min, 0.18-0.6 mg/kg/h). Understandably, some patients may be benefiting due to a placebo effect especially with the use of a “new” intervention. Yet, overall, unlike adult patients who may not tolerate LDKI well leading to lack of consensus on benefit, the limited studies in pediatric patients do not show high rates of side effects leading to intolerability or discontinuation, parallel to what we show here. [29-31] Thus, similar to the conclusion of Bredlau et al. [31], we believe that LDKI is well tolerated and may be a viable option for refractory pediatric cancer- and chronic sickle cell-related pain and therefore should be offered routinely for these indications and assessed for benefit. Future study is required to further delineate appropriate dosing and the pediatric hematology/oncology patients most likely to benefit as well as to explore utilizing LDKI in the outpatient setting.