Comparing Acute Pain Management Protocols for Patients With Sickle Cell Disease

National Institutes of Health (NIH), National Heart, Lung, and Blood Institute (NHLBI), University of Cincinnati, Mount Sinai Hospital, New York

The goal of this pilot study is to improve emergency department (ED) pain management for adults with sickle cell disease. Sickle cell disease (SCD) is the most common genetic disorder in the United States, and occurs primarily among African Americans. Management of painful episodes associated with SCD, referred to as vaso-occlusive crises (VOC), is the most common reason for SCD patients to visit the ED. Currently, there is no standard approach to managing VOC pain in the ED that is widely accepted and used, and pain management for vaso-occlusive crisis in persons with SCD is very different between providers and not based on research. Many times, patients who come to the ED with sickle cell pain feel that they do not receive adequate pain control. If EDs could provide efficient, effective, safe, patient-centered analgesic management, it may be possible to improve pain management for adults with SCD experiencing a VOC. Guidelines for treating vaso-occlusive crises caused by sickle cell disease will soon be published by the National Heart, Lung and Blood Institute of the National Institutes of Health. These guidelines recommend patient-specific pain treatment protocols or a standardized pain management protocol for SCD when a patient does not already have a pain treatment protocol designed for them. The purpose of this pilot study is to compare these two ways to treat vaso-occlusive pain in the ED for adults with sickle cell disease, and to determine if a large randomized controlled trial is feasible and required.

In August 2012 the National Heart, Lung, and Blood Institute (NHBLI) released for public comment their "Management of Sickle Cell Disease" evidence-based recommendations that were developed with consensus panel expertise. Because of a lack of empirical data, most of the recommendations specific to vaso-occlusive crises (VOC) were based on consensus panel expertise. Recommendations included the use of a patient-specific protocol (specific agents and doses for an individual patient). While many attempts have been made to implement patient-specific analgesic protocols for use in emergency departments (EDs), anecdotally, these have been difficult to implement and maintain over time; a practical approach to development, implementation, and dissemination has not been determined. As patient-specific protocols are not available in most EDs, the guidelines go on to recommend a SCD specific standard analgesic protocol. Both of these recommended protocols provide more aggressive VOC pain management than a typical generic ED pain protocol. However, there is an urgent need to rigorously test the NHLBI recommendations and compare the two approaches for managing VOC in the ED. A large randomized clinical trial (RCT) is essential to test these protocols. This pilot project will compare these two different, evidence-based, protocols which include opioid pain medicines routinely used as standard of care to treat VOC pain in the ED for individuals with SCD, and collect the data necessary to determine if a large RCT is feasible and required. This study is novel in that it will design an approach to develop and implement patient-specific and standard analgesic VOC protocols for use in the ED, will develop a bundle of information technology and education interventions to enhance protocol adoption for the pilot RCT, and also be the first RCT conducted in an ED setting to compare two different ED pain management protocols for SCD patients who experience a VOC. The study consists of 3 aims: Develop and implement patient specific VOC protocols for patients randomized to this arm, Conduct a pilot RCT to determine the necessary sample size needed for a large RCT to compare the difference in reduction in pain score from ED arrival to discharge, hospitalization, clinical and safety outcomes, between subjects assigned randomly to either a standard SCD analgesic protocol or to a patient-specific analgesic protocol, Measure feasibility of methods and acceptability of and fidelity to protocols by evaluating optimal recruitment and retention strategies, and assessing ED providers perceptions of facilitators and barriers to protocol use and protocol adherence. The soon to be published NHBLI guidelines for managing SCD will be used as the standard protocol with the modification of basing the initial dose of pain medicine on patient weight. The standard protocol will recommend re-assessment, and re-dosing with possible dose escalation, every 20-30 minutes. Repeat doses for patients randomized to the weight-based protocol, when necessary, will be maintained or provided at 1 dose level increase (no more than 25%) above the initial dose. For patients randomized to the patient -specific protocol, the SCD provider has experience with the individual patient and is best qualified to make dosing and frequency recommendations based upon doses required during past ED and hospital visits for treatment of VOC, and on daily opioid use if applicable. There is no set maximum dose for patients randomized to the patient-specific protocols. Both the patient-specific and standard protocols will be available in the ED via a patient's electronic medical record. Upon ED arrival, providers will retrieve the patient's study protocol (patient-specific, or standard) which will include the starting agents, and doses, the subsequent analgesic recommendations, and order medications according to the pre-determined protocol. The study will be conducted at the emergency departments of the University of Cincinnati Medical Center and the Mt. Sinai Hospital in New York. Patients will be enrolled in the study for up to 12-months, but may contribute no more that five different ED visits for VOC pain control during enrollment to allow for a larger number of different patients. Study outcomes will be compared between ED visits of patients randomized to a patient-specific vs. a standard SCD protocol. The primary outcome will be the difference in pain score from ED arrival to discharge, up to 6 hours, as measured using a visual analogue scale. The trajectories of average pain scores from immediately prior to administration of 1st analgesic dose to discharge by 30 minute increments for each treatment group will also be calculated.

A patient-specific analgesic protocol for use in the ED to manage VOC crises. Following randomization, a patient's healthcare team will develop a specific analgesic protocol for use during future ED visits for VOC occurring during the study period (up to 5 visits). Treatment protocols will include either morphine sulfate or hydromorphone (delivered intravenous or sub-cutaneous). Dosage and frequency will be based on a patient's prior treatment history.

A standardized analgesic protocol (based on recent NHLBI recommendations) for use in the ED to manage VOC crises. Treatment protocol will include either morphine sulfate or hydromorphone (delivered intravenous or sub-cutaneous), with dosage based on weight. Repeat doses of opioids may be administered every 20-30 minutes as needed, although dosage will be maintained or provided at no more than 25% above the initial dose.

Standardized analgesic management using a SCD specific standard protocol based on NHBLI guidelines (initial opioid dose weight-based).

Standardized analgesic management using a SCD specific standard protocol based on NHBLI guidelines (initial opioid dose weight-based).

Patient specific analgesic management, with specific opioid dosage and frequency based on a protocol developed by a patient's healthcare team.

Patient specific analgesic management, with specific opioid dosage and frequency based on a protocol developed by a patient's healthcare team.

Each ED study visit was the unit of analysis for the statistical methods addressing the primary outcome. The primary outcome was change in pain score from arrival to discharge. Pain severity was assessed at arrival and discharge from ED using a 100 mm visual analogue scale (VAS). The VAS range is 0 to 100 with 0 indicating "no pain" and 100 indicating "pain as bad as it could be" or "worst imaginable pain".Discharge was defined by which one of the following occurred first: (a) decision to admit to hospital; (b) patient physically leaves the ED to home; or (c) after six hours of observation in the ED. Thus, the difference in pain scores were calculated as the arrival minus discharge VAS scores, with higher positive pain difference or change scores indicating greater pain reduction.

Pain severity was assessed at arrival and every 30 minutes until discharge from the ED using a 100 mm visual analogue scale (VAS). The VAS range is 0 to 100 with 0 indicating "no pain" and 100 indicating "pain as bad as it could be" or "worst imaginable pain". Discharge was defined by which one of the following occurred first: (a) decision to admit to hospital; (b) patient physically leaves the ED to home; or (c) after six hours of observation in the ED. A hierarchical random coefficients regression model for repeated measurements (type of mixed hierarchical mixed-effect model) was conducted on the pain scores collected at six time points (arrival, post-placement 30-min, 60-min, 90-min,120-min, discharge) to evaluate the trajectory of change in pain. Discharge occurred at 120 minutes or later during each visit, with the exception of one discharge at 54 minutes.

Nausea at any point from placement until discharge, based on nausea data collected every 30 minutes during that time period. Thus, a nausea variable was derived in which 0=no and 1=yes that nausea was reported by the patient at least once during the placement to discharge time interval.

Vomiting at any point from placement until discharge, based on vomiting data collected every 30 minutes during that time period. Thus, a vomiting variable was derived in which 0=no and 1=yes that vomiting was reported by the patient at least once during the placement to discharge time interval.

Incidence of a Decrease in Systolic Blood Pressure Greater Than or Equal to 20% of Baseline During Emergency Department Visit

Decrease in systolic blood pressure at any point from placement until discharge, based on blood pressure data collected every 30 minutes during that time period. A systolic variable was derived in which 0=no and 1=yes that a >= 20% decrease of baseline systolic blood pressure was reported by the patient at least once during the placement to discharge time interval.

Incidence of a Decrease in Diastolic Blood Pressure Greater Than or Equal to 20% of Baseline During Emergency Department Visit

Decrease in diastolic blood pressure at any point from placement until discharge, based on blood pressure data collected every 30 minutes during that time period. A diastolic variable was derived in which 0=no and 1=yes that a > 20% decrease of baseline diastolic blood pressure was reported by the patient at least once during the placement to discharge time interval.

Saturation of peripheral capillary oxygen < 95% (SPO2 < 95%) at any point from placement until discharge, based on SPO2 data collected every 30 minutes during that time period. Thus, a SPO2 variable was derived in which 0=no and 1=yes that SPO2 < 95% was reported by the patient at least once during the placement to discharge time interval.

Respiratory distress at any point from placement until discharge, based on data collected every 30 minutes during that time period. Thus, a respiratory distress variable was derived in which 0=no and 1=yes that respiratory distress was reported by the patient at least once during the placement to discharge time interval.

Severe-to moderate sedation at any point from placement until discharge, based on sedation data collected every 30 minutes during that time period. Thus, a sedation variable was derived in which 0=no and 1=yes that moderate-severe sedation was reported by the patient at least once during the placement to discharge time interval. Sedations scoring was as follows: None was defined as "awake and alert", Mild sedation was defined as "responds to voice", Moderate sedation was defined as "responds to touch, with or without voice" and Severe sedation was defined as "somnolent, difficult to arouse".

Intubation or other assistive ventilation techniques - including bag, valve, or mask was performed during the ED stay; this was determined at discharge.

Inclusion Criteria: Adult SCD patients with genotypes SS, SC, SB+, or SB- Exclusion Criteria: Patients with sickle cell trait Allergic to both morphine sulfate and hydromorphone, Patients who have an explicit care plan that states they cannot be admitted to the hospital for pain control, Non-English speaking, Patients admitted for a medical complication, Record of >24 ED visits in the prior 12 months, Children