a b s t r a c t

Importance: Studies have demonstrated the benefits of INF in reducing pain scores in pediatric patients with VOC due to Sickle cell disease (SCD) and in adult patients with chronic pain conditions other than VOC, such as cancer. However, there is limited literature that exists describing the role of INF in adult patients with VOC due to SCD. Current literature demonstrates that the use of IV morphine for VOC patients leads to reduced pain. Therefore, comparing the use of INF with IV morphine will establish the degree of effectiveness of INF for VOC patients.

Objective: To determine if Intranasal fentanyl is equally as effective as IV morphine for treating VOC-associated pain in adult SCD patients.

Design: This study was a retrospective non-inferiority cohort study. electronic health records were utilized to identify eligible patients between January 1, 2021 to February 28, 2022. Patients who received INF as an initial opioid upon presentation to the ED where allocated to the intervention group. On the other hand, individuals who received IV morphine as an initial opioid upon presentation to the ED were allocated to the control group. Setting: A multi-site healthcare system containing five hospitals.

Participants: Patients 18 years of age or older, admitted to the ED with VOC due to SCD, and received INF or IV

morphine as an initial opioid upon presentation to the ED.

Main outcomes and measures: The primary outcome was to evaluate the percent change in pain reduction after the initial dose of opiate between groups. Secondary outcomes include time to first Rescue medication, total mor- phine milligram equivalent (MME) of IV opiates, hypotension, bradycardia, respiratory distress requiring opiate reversal within 6 h post- study drug administration, readmission within 48 h, and ED disposition.

Results: A total of 230 patients were reviewed within the study period, 95 subjects met inclusion criteria, 31 sub- jects were included in the INF arm and 64 subjects in the IV morphine arm. The primary outcome showed an av- erage percent pain reduction of 17.25% in the INF arm and 17.15% in the IV morphine arm. The point estimate difference was 0.1% (95% CI -9.3%-9.5%; non-inferiority (p < 0.0001). The median dose of IV opiates was 8 MME in the INF group, and 6 MME in the IV morphine group (p = 0.0268). The time from study drug to first res- cue medication administration was 22.4 min and 27.3 min in the INF and IV morphine groups, respectively (p = 0.2231). There was no incidence of hypotension or respiratory distress requiring opiate reversal in either arm. Bradycardia occurred in 12.9% and 7.7% (p = 0.2042), readmission rates within 48 h due to VOC was 6.5% and 20.9% (p = 0.0553), and discharge from the ED to home was 16% and 66% (p = 0.0196) in INF and IV morphine arms, respectively.

Conclusion: INF provided similar pain reduction compared to IV morphine in the treatment of adults with VOC presenting to the ED. IV morphine arm showed a statistically significant difference in discharge to home from the ED, however there was a trend in readmission within 48 h. The study showed no significant difference in hy- potension, respiratory distress, or bradycardia between the groups. The INF group had no significant impact on time to drug administration compared to IV morphine, however it was within 1 h of patient presentation which complies with American Society of Hematology (ASH) guidelines. In conclusion, our study showed that

* Corresponding author.

E-mail addresses: [email protected] (O. Assad), [email protected] (R. Zamora), [email protected] (K. Brown), [email protected] (L. Melnitsky), [email protected] (J. Moses), [email protected] (V. Sherman).

https://doi.org/10.1016/j.ajem.2022.11.026

0735-6757/(C) 2022

INF was non-inferior when compared to IV morphine in reducing pain scores after drug administration. There- fore, INF is an effective alternative to IV morphine for pain management in adults presenting to the ED for VOC particularly in those with limited IV access.

(C) 2022

1. Introduction

Sickle cell disease is an inherited red blood cell disorder re- sulting from a genetic mutation that encodes the hemoglobin protein [1]. This mutation causes the red blood cell (RBC) to become inflexible with a concave sickle shape. The misshaped RBCs are unable to trans- port oxygen effectively as they adhere to one another, which leads to the blockade of the small blood vessels causing tissue ischemia, necrosis, and sudden pain crisis, which known as vaso-occlusive crisis (VOC) [1]. The cells that cause VOC not only consist of misshaped RBCs but also in- flammatory cells. Hence, inflammation seems to play a major role in the development of VOC along with infections, dehydration, and cold weather [13]. VOC flare ups commonly occurs in the lower back, legs, hips, abdomen, and chest, which could lead to foot ulcers, stroke, and kidney insufficiency. There are various medications utilized to prevent flare ups associated with SCD, however these may not always be effec- tive. This may lead to patients requiring medical attention at an ED or clinic. Approximately 70% of SCD ED admissions are due to uncontrolled pain associated with VOC [2]. Therefore, per the ASH guidelines, control- ling the pain associated with VOC by providing appropriate analgesia within 60 min of arrival to the ED is a crucial step in SCD treatment [3]. VOC management is guided by the patient’s self-reported pain sever-

ity using a Pain severity chart with the goal of providing quick and effec- tive pain management often achieved with IV opioids such as morphine. Limitations in achieving adequate pain control include Emergency Department (ED) wait times and obtaining Intravenous access. Identi- fying alternative pain control methods potentially leads to more rapid pain relief and improved patient satisfaction, thereby decreasing total time in the ED. A prior study involving the pediatric population with SCD found that there was a significant pain reduction at 20 min when comparing in- tranasal fentanyl (INF) to placebo [2]. Furthermore, another study which also included the pediatric population found that there was no significant difference in pain reduction when comparing INF to IV analgesia [12].

Some appealing aspects of INF are that it is a potent opioid with high bioavailability and leads to less histamine release, which could contribute to the lower rates of hemodynamic instabilities compared to the Large doses of IV morphine [4]. Drug administration through the intranasal route is quick, painless, convenient, and can be delivered with minimal training. Furthermore, the IN route may allow for more rapid pain relief and higher patient and provider satisfaction when compared to other routes of administration [5]. The literature has demonstrated that the use of INF in an order set has improved time to first opioid dose [11]. In addition, a meta-analysis involving the cancer population supported the non-inferiority of INF versus IV opioids with the advantage of shorter Time to administration [7]. That being said, there is limited literature reviewing the use of INF in the adult VOC patient population. Therefore, INF was added as a treatment option for adult SCD patients presenting to the ED with the intention of providing quick and effective care. The purpose of this study is to evaluate the analgesic effect of INF in SCD patients presenting with VOC compared to IV morphine.

1. Methods
   1. Study design and setting

This was a multi-site, retrospective, non-inferiority cohort study conducted at a large healthcare system. Our healthcare system consists

of five adult emergency departments providing care to patients with VOC due to SCD. This study was deemed exempt by the Institutional Review Board.

• 1. Data source and study population

A report was generated to identify patients that received either IV morphine or INF in the ED during the study period from January1, 2021 to February28, 2022. Outpatient opiate regimens were collected via a statewide Prescription Drug Monitoring Program to determine the baseline opiate use for patients within both arms. All identified pa- tients were reviewed to determine eligibility. The study included pa- tients who were 18 years of age or older, admitted to the ED with VOC due to SCD, and received INF or IV morphine as an initial opioid upon presentation to the ED. The study excluded patients who did not receive INF or IV morphine as the initial opioid, admitted to the ICU with Acute chest syndrome, had a known allergy or contraindication to opiates, and had bilateral occluded nasal passages or epistaxis. Patients who received INF as an initial opioid upon presentation to the ED were allocated to the intervention group. Patients who received IV morphine as an initial opi- oid upon presentation to the ED were allocated to the control group.

• 1. Intervention

With the intention to provide quick and effective care, INF was added as a treatment option for adult SCD patients presenting to the ED. The order set includes INF with either a 50 ug or 100 ug dose. The cal- culated initial median dose in the INF arm was 100 ug. The order set also includes IV morphine as an alternative option at a 0.1 mg/kg dose. The calculated initial median dose in the IV morphine arm was 6.4 mg.

• 1. Outcome measures

The primary outcome of this study was to compare the percent change in self-reported pain scores in patients admitted to the ED with acute pain crisis due to SCD and treated with INF or IV morphine. This outcome was assessed based on the difference between the pain score prior to the study drug administration and the pain score post- study drug administration. The secondary outcomes were the time to first rescue medication, total morphine milligram equivalent (MME) of IV opiates, hypotension defined as systolic blood pressure < 90 mmHg, bradycardia defined as a heart rate < 60 beats per minutes, re- spiratory distress requiring opiate reversal within 6 h post-study drug administration, readmission within 48 h due to pain crisis, and ED dis- position to home.

• 1. Statistical analysis

Based on available literature and population sampling, a sample size of 93 at a 1 to 2 ratio is required to achieve 80% power to detect non- inferiority of INF versus IV morphine using a one-sided, two sample t– test, with a non-inferiority margin of 25% and alpha of 0.025. The predefined non-inferiority margin was selected based on the findings of Farrar et al. where a clinically significant pain relief reduction in a nu- merical pain score was 30%. Descriptive statistics were applied to test for differences in clinical characteristics between control and interven- tion group. Continuous data was analyzed using means via two sample

Patient characteristics.

Characteristics	INF	IV Morphine	p-value
Age, yr, mean (SD)	31.1 (10.4)	31.8 (9.2)	0.5059
Weight kg, mean (SD)	72.9 (17.8)	69.3 (16.2)	0.2576
Hemoglobin g/dl, mean (SD)	8.8 (2.1)	8.4 (1.9)	0.5064
Received drugs that inhibits Cytochrome P450 isoenzyme in the last 30 days, n (%)	0 (0)	1 (1.6)	0.6737
Acute bronchial asthma/upper airway obstruction, n (%)	4 (12.9)	5 (7.8)	0.2042
Nonsteroidal anti-inflammatory drugs within 24 h before the first dose of study drugs, n (%)	13 (42)	19 (29.7)	0.2363
IV hydration, n (%)	31 (100)	64 (100)	1
Fill history of hydroxyurea, L-glutamine, crizanlizumab, voxelotor, n (%)	8 (25.8)	15 (23.4)	0.6487
Initial dose, median (IQR)	100 (62.5; 100)	6.4 (5.6; 7.2)	–
Time from study drug to first pain score in minutes, mean (SD)	26.7 (15.6)	27.3 (14.8)	0.8665
Daily MME per E-FORCE, median (IQR)	82.5 (55.4-150)	120 (67.5-228.6)	0.1152

independent t-test or paired t-test. Non-continuous data was analyzed using non-parametric tests.

1. Results
   1. Patient characteristics

A total of 230 patients were reviewed within the study period. Of the 230 patients, 135 were excluded with 35 patients who did not receive the study drugs, 60 patients had a diagnosis other than SCD (trauma), 15 patients with allergies to opiates, 15 patients who received both study drugs at the same time, and 10 patients without documented pain scores either prior to the study drug administration or post-study drug administration. Therefore, 95 subjects met inclusion criteria with 31 subjects allocated to the INF arm and 64 subjects to the IV morphine arm. There were no statistical differences in baseline characteristics between both arms (Table 1)

• 1. Primary outcome

The primary outcome showed an average percent pain reduction of 17.25% in the INF arm and 17.15% in the IV morphine arm. The point es- timate difference was 0.1% (CI: 95%; -9.3%-9.5% non-inferiority (p < 0.0001)).

• 1. Secondary outcome

The median IV opiates dose was 8 MME in the INF group and 6 MME in the IV morphine group (p = 0.0268). The time from study drug to first rescue medication administration was 22.4 min and 27.3 min in the INF and IV morphine groups, respectively (p = 0.2231). There was no incidence of hypotension or respiratory distress requiring opiate re- versal in either group, however bradycardia occurred in 12.9% and 7.7% (p = 0.2042) of patients in the INF and IV morphine groups, respec- tively. readmission rates within 48 h due to VOC were 6.5% in the INF and 20.9% in the IV morphine groups (p = 0.0553) and 16% and 66% of patients were discharged directly from the ED (p = 0.0196) in the INF and IV morphine arms, respectively (See Table 2).

1. Discussion

Patients who presented to the emergency department (ED) as a result of vaso-occlusive crisis (VOC) were reviewed in this multi-site, retrospective cohort study. The study results showed that INF was non-inferior to IV morphine when comparing pain score reduction after study drug administration. The point estimate difference between study groups was 0.1%, which was well below the predefined non- inferiority margin with a (p < 0.0001). Therefore, these findings sup- ports the use of INF as an effective alternative to IV morphine for pain management in adults SCD patients presenting to the ED with VOC.

There are approximately 100,000 individuals living with SCD in the United States. Many of these persons seek care at an emergency depart- ment at some point. One study showed that at least half of those with SCD in the state of California sought care at least once at an ED between the years of 2005 and 2014 [8]. Pain is the leading cause of emergency department visits in patients with SCD and thus the mainstay of treat- ment is pain control, often achieved by IV opioids [9,10]. As mentioned earlier, some of the limitations in achieving adequate pain control include ED wait times and obtaining vascular access, which poses inher- ent risks and occasional challenges. A prior study involving the pediatric population found that there was no significant difference in pain trajec- tories or changes in pain scores when comparing INF to “routine care” comprised of IV analgesia [12]. This is consistent with our findings that using intranasal fentanyl is non-inferior to IV morphine when mea- suring pain reduction.

One attractive aspect of INF is its potential to be administered quicker than an IV alternative. Prior research has shown that utilizing INF in an order set has an advantage by reducing time to first opioid dose administration [11]. This is contradictory to our findings as there was no significant improvement in the time to the first-dose opioid ad- ministration. Importantly, the same study found that time to first opioid administration, whether INF or IV opioids, made no significant differ- ence in disposition outcomes. We found that those who initially re- ceived INF were less likely to be discharged home. Furthermore, the study showed no difference in hypotension, respiratory distress, or bra- dycardia between the groups. This is consistent with our study findings. Our study had several strengths which includes that power was met to show non-inferiority, no statistical differences in baseline character- istics between groups, and its multi-site setting. We acknowledge that various limitations exist in our study. Of notable importance is that we cannot address why some individuals initially received IV morphine while others received INF. It is possible that these groups of individuals varied in such a way that was not captured by this study. For example, individuals with medication ports can be presumed to have received IV morphine more often than INF, while patients with difficult-to- obtain IV access would most often have received INF. This study was a retrospective chart review relying solely on user documentation. Pain score reporting was not done at a standard interval and was depended on ED workflow. Although power was achieved, the small sample size may limit external validity. We were unable to capture patients that presented to the ED with high morphine requirements and/or limited IV access. Additionally, our study did not take into consideration precip- itating factors that lead to VOC which could impact the pain severity and disposition outcomes. It is possible that capturing the precipitating fac- tor at baseline may explain the significantly higher rates of admission in the INF arm compared to the IV morphine arm. For example, patients found to have an infection may have a higher chance of being admitted for antibiotics compared to those patients with dehydration as the pre- cipitating factor. Furthermore, patients in both the INF and IV morphine arms received IV morphine as a rescue medication. This may have had a greater influence than the initial study medication administration on

Primary and Secondary outcomes.

Primary endpoint	INF	IV morphine	p-value
Average percent pain reduction	17.25%	17.15%	<0.0001 (non-Inferiority)
Secondary endpoint
Time from triage to study drug administration in minute, mean (SD)	29.2 (16.4)	29.7 (15.6)	0.8732
Time from study drug to first rescue medication in minutes, mean (SD)	22.4 (17.3)	27.3 (18.5)	0.2231
Total MME of IV opiates within 6 h of-study drug administration, median (IQR)	8 (6-11.4)	6 (5.7-9.3)	0.0268
Incidence of hypotension within 6 h of study drug administration, n (%)	0 (0)	0 (0)	1
Bradycardia within 6 h of study drug administration, n (%)	4 (12.9)	5 (7.7)	0.2042
Respiratory distress requiring opiate reversal within 6 h of study drug administration, n (%)	0 (0)	0 (0)	1
Readmission within 48 h, n (%)	2 (6.5)	13 (20.9)	0.0553
ED disposition home, n (%)	5 (16)	41 (66)	0.0196

secondary outcomes. Finally, the use of INF was a new practice in our system which may have delayed time to administration due to unfamil- iarity with its use.

In our study, we have shown that INF is just as effective as IV mor- phine. Future research should be directed at studying if the use of INF can decrease time to disposition and decrease time to pain control. A controlled trial between INF and IV morphine may also provide further insight into the efficacy of these two drugs in the setting of acute sickle cell pain crises. While this study examined the appropriate use of opioid medications in a clinically indicated condition, it is important to ac- knowledge the crisis for which this class of drugs is implicated. Further studies can examine the use of non-opioid medications in achieving our primary and secondary outcomes.

Based on the available data, INF is an appropriate alternative to IV morphine as a first-line analgesic agent in patients presenting to the ED for vaso-occlusive sickle cell pain crises. The ability to administer ad- equate analgesia before obtaining vascular access can play a role in im- proving patient satisfaction and play a larger role in the overall efficiency in the care of these patients.

1. Conclusion

INF provided similar pain reduction compared to IV morphine in the treatment of adults with VOC presenting to the ED. IV morphine arm showed a statistically significant difference in discharge to home from the ED, however there was a trend in readmission within 48 h. The study showed no significant difference in hypotension, respiratory dis- tress, or bradycardia between the groups. This may have resulted from the small sample size of our study where patients who require higher doses of IV morphine leading to increased adverse effects were not cap- tured. There was no significant difference in time to study drug adminis- tration between groups, however medications were given within 1 h of patient presentation. In conclusion, our study showed that INF was non- inferior when compared to IV morphine as demonstrated by the change in pain scores prior to and after drug administration. Therefore, INF is an effective alternative to IV morphine for pain management in adults pre- senting to the ED for VOC particularly in those with limited IV access.

CRediT authorship contribution statement

Osama Assad: Writing – original draft, Methodology, Formal analy- sis, Data curation, Conceptualization. Rolando Zamora: Writing – original draft, Formal analysis, Data curation. Kyle Brown: Writing –

review & editing, Supervision, Resources, Project administration, Meth- odology, Formal analysis, Data curation, Conceptualization. Leon Melnitsky: Supervision, Resources, Methodology, Conceptualization. Jessica Moses: Writing – review & editing, Supervision, Project admin- istration, Methodology, Formal analysis, Data curation, Conceptualiza- tion. Veronica Sherman: Writing – review & editing, Supervision, Resources, Project administration, Methodology, Formal analysis, Data curation, Conceptualization.

References

1. Inusa BPD, Hsu LL, Kohli N, et al. Sickle cell disease-genetics, pathophysiology, clin- ical presentation and treatment. Int J Neonatal Screen. 2019;5(2):20. Published 2019 May 7. https://doi.org/10.3390/ijns5020020.
2. Barrett MJ, Cronin J, Murphy A, et al. Intranasal fentanyl versus intravenous mor- phine in the emergency department treatment of severe painful sickle cell crises in children: study protocol for a randomised controlled trial. Trials. 2012;13:74.
3. Yawn BP, Buchanan GR, Afenyi-Annan AN, et al. Management of sickle cell disease: summary of the 2014 evidence-based report by expert panel members [published correction appears in JAMA. 2014 Nov 12;312(18):1932] [published correction appears in JAMA. 2015 Feb 17;313(7):729]. JAMA. 2014;312(10):1033-48. https://

doi.org/10.1001/jama.2014.10517.

1. Farrar JT, Young Jr JP, LaMoreaux L, Werth JL, Poole MR. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain. 2001;94(2):149-58. https://doi.org/10.1016/S0304-3959(01)00349-9.
2. Dale O, Hjortkjaer R, Kharasch ED. Nasal administration of opioids for pain manage- ment in adults. Acta Anaesthesiol Scand. 2002;46:759-70. https://doi.org/10.1034/j. 1399-6576.2002.460702.
3. Fein DM, Avner JR, Scharbach K, Manwani D, Khine H. Intranasal fentanyl for initial treatment of vaso-occlusive crisis in sickle cell disease. Pediatr Blood Cancer. 2017;

64(6). https://doi.org/10.1002/pbc.26332.

1. Giroir BP, Collins F. The state of sickle cell disease care in the United States: how can emergency medicine contribute? Ann Emerg Med. 2020;76(3S):S1-3. https://doi. org/10.1016/j.annemergmed.2020.07.029.
2. Glassberg JA. Improving emergency department-based care of sickle cell pain. Hematology Am Soc Hematol Educ Program. 2017;2017(1):412-7. https://doi.org/ 10.1182/asheducation-2017.1.412.
3. Lovett PB, Sule HP, Lopez BL. Sickle cell disease in the emergency department. Hematol Oncol Clin North Am. 2017;31(6):1061-79. https://doi.org/10.1016/j.hoc. 2017.08.009.
4. Kelly GS, Stewart RW, Strouse JJ, Anders JF. Intranasal fentanyl improves time to analgesic delivery in sickle cell pain crises. Am J Emerg Med. 2018;36(7):1305-7. https://doi.org/10.1016/j.ajem.2017.11.015.
5. Myrick R, Blakemore S, Waite E, et al. Outpatient pain clinic and intranasal fentanyl to improve sickle cell Disease outcomes. Pediatr Blood Cancer. 2020;67(10):e28648. https://doi.org/10.1002/pbc.28648.
6. Ahmed SG. The role of infection in the pathogenesis of vaso-occlusive crisis in pa- tients with sickle cell disease. Mediterr J Hematol Infect Dis. 2011;3(1):e2011028. https://doi.org/10.4084/MJHID.2011.028. Epub 2011 Jul 8. PMID: 21869914; PMCID: PMC3152450.